426

CHAPTER - V

CONCLUSIONS

The Gas Insulated System (GIS) is a reliable system for electrical

energy transport. One of the important parts of the GIS is the insulating gas

that can be stressed with lightning impulses, if a busduct connector is being

closed or opened. First, SF6 was used as gas insulation. Interest in electrical

insulating properties of compressed gases and gas mixtures has existed for

many years and is currently stimulated by the desire to alternatives to SF6

gas. This gas is very expensive and non-friendly for the environment

because of its high global warming potential, but it possesses very good

dielectric properties relative to some other gases. An alternative gaseous

medium is needed in order to substitute completely or at least partially the

expensive gas SF6. This aspect is also important with respect to

environmental considerations (greenhouse effect). As an alternative, N2/SF6

mixtures are chosen. SF6 / N2 gas mixtures showed a good synergy effect

even with only 10 % admixture of SF6 to Nitrogen. For further optimization

and improvement of the gas insulation, some simulation trials have been

carried out and the results are presented and discussed.

In this regard, uniform field breakdown data are of considerable

interest due to both theoretical and practical considerations. From the

theoretical point of view such data are important as they provide valuable
 427

information on the fundamental physical processes. Practically, the data

form the basis of design criteria for compressed gas insulated systems.

The Components and Construction of Gas Insulated Substations and

major technological developments that have taken place in the field of Gas

Insulated Substations that are relevant to the work reported in this thesis

are presented. Defects occurring in Gas Insulated Substations and the

methods for their control and deactivation are reviewed. An uncharged

metallic particle resting on bare electrode in a Gas Insulated System will

gradually acquire charge due to the application of electric field around it.

The charge accumulated is a function of Electric field, shape, size and

orientation of the particle. Three forces act on the particle: eg. Electrostatic

force and oppositely acting gravitational and drag forces. When electrostatic

force exceeds the gravitational and drag forces the particle lifts from its

position. A further increase in the applied voltage makes the particle move

into the inter electrode gap in the direction of applied field. This increases

the probability of a flashover. If an investigation reveals the presence of a

particle, it is required to analyze the particle i.e., find the material and

approximate size of the particle as certain particles are more deleterious

than others (For example for a 100kV Aluminium particle of Size 10mm in

length and 0.25 as radius with a 152/55 Bare enclosure the movement is

24.24214 mm, whereas for copper and Silver for the identical condition it is

5.054871 mm and 3.937674. Thus, the extent of movement in the radial

direction can be used as a criterion for identifying the type of particle.

The influence of increased voltage level on the motion of the particles

is also investigated. If the calculations, as described above, are performed at

a higher voltage level, the particle will lift higher from the surface and the

time between bounces will increase (For example: Aluminium Particle of

size 10mm/0.25 radius on a 152/55 Bare enclosure is applied with 100kV

the maximum movement is 24.2421 mm and for the same condition with

applied voltage of 145kV it is found to be 31.50553mm).

The work reported in this thesis deals with the movement of the

metallic particles in a 3-phase common enclosure busduct and single-phase

isolated conductor GIS busduct with bare electrodes as well as coated

electrodes under pure SF6 gas and also with SF6+N2 gas mixtures. The

macroscopic electric field at the surface of the enclosure for the 3-phase

system is calculated in Cartesian coordinates. The electric field has been

used to determine the charge as well as the force on the particle. The radial

movement is calculated using the standard equation of motion. The

calculations have been done for power frequency voltages. The results

obtained from the calculations show that additional information about the

particle could be obtained when voltage dependence is introduced in the

calculations. For instance, it can be noted that aluminum particles (Max.

Movement for 500kV is 36.48855mm) are more influenced by the voltage

than copper (Max. Movement for 500kV is 13.29363mm) and silver particles

(Max. Movement for 500kV is 11.09344) due to its lighter mass. This results

in the aluminum particle acquiring greater charge-to-mass ratio and hence

attaining maximum movement when compared to copper and silver

particles.

There is an increasing interest in the applications of mixture of SF 6

and other common gases. Mixtures are suitable particularly for applications

where low ambient temperatures could result in liquefying SF6 if operated at

higher pressures. The use of a mixture of SF6 and a cheap inert gas could

eliminate some of the problems associated with pure SF 6 and reduce the

insulation cost. Studies have shown that SF6 gas mixtures can have equally

good overall electrical properties as compared to pure SF6. Maximum

movement is computed with SF6 gas mixtures by changing the Reynold’s

number and viscosity in the simulation.

In this thesis simulations on particle movement have been carried out

under pure SF6 and also with SF6+N2 gas mixtures. Sulphur hexafluoride

(SF6) is widely used gas for applications in power system due to its high

dielectric strength and good arc interruption properties. However, SF6 gas

has been found to be a green house gas that causes global warming. Among

the environmentally benign gases, alternative to SF6 gas, the SF6- N2 gas

mixtures is regarded as one of the most attractive gases for the same setup

used for SF6, because of the synergetic effect in electrical insulation

performance, thus reliability of GIS can be improved. Simulation considers

the effect of electric field on the movement of metallic contaminant in pure

SF6 gas and mixture of gases (SF6/N2) in different proportions and results

have been presented and analyzed It is found that SF6+N2 mixture of gases
 430

perform better when compared to pure SF6 gas in the presence of metallic

particle contamination (for example a Copper contaminant under pure SF6

gas environment attains maximum radial movement of 5.054871,whereas

with 40%,50% and 60% of N2 in gas mixtures the movement of the particle

is obtained as 4.457769,4.210333 and 4.492601 respectively ). Generally

40%-60% of N2 in SF6+N2 gas mixtures generally referred as optimal mixture

of gases which can substitute for pure SF6 for the same set up electrical

equipment in GIS and GITL.

The above investigations have been carried out in the case of single

phase isolated conductor for different types of voltages eg. Power frequency,

switching and lightning impulses super imposed on power frequency, axial

movement of the Al, Cu and Ag particles has been determined using Monte-

Carlo simulation technique. This technique has been used for isolated

conductor and common enclosure three-phase busduct as well as coated

and uncoated systems. Under coated condition and in the case of three

phase common enclosure gas insulated busduct investigations are carried

out for power frequency voltages.

Simulations on particle movement have been conducted under AC

field stress. Particle dimensions and voltage levels as well as gas pressure

have been chosen in accordance with the design. The coefficient of

restitution, which denotes the ratio of outgoing to incoming velocities, is of

vital importance for determining the maximum movement of particle. The

primary goal of simulation is to create a satisfactory mathematical model of

the metallic particle motion in the GIS under pure SF6 and also with SF6+N2

gas mixtures.

Dielectric coating applied to the inner surface of the outer enclosure of

a coaxial GIS / GITL system improves the insulation performance in several

ways. Coating has the effect of smoothing the electrode surface and

reducing the pre-breakdown current in the gas gap. Also, the electrostatic

charge acquired by a particle is reduced and hence the range of its motion

under an applied power frequency field is inhibited (Typically for 152/55

Dia. Uncoated Electrode the maximum movement for 200kV Aluminum

particle is 61.1385mm.whereas for the coated electrode with 50 m

thickness it is found to be 0.795778mm, 0.651594mm with 75 m

thickness and 0.291621mm with 200 m thickness coating. The above

investigations with coated electrode conditions are also carried out with SF6

+N2 gas mixtures for 200m thickness and the movement of the particle

being very small with coating of 200m thickness given to the enclosure, the

change in viscosity with change in concentration of N2 in gas mixtures do

not have much impact on the movement of the metallic particle. Thus

movement of metallic particle can be restricted under SF6 and also with

SF6+N2 gas mixtures using dielectric coated electrodes.

The present work also deals with the presence of contaminant

on the spacer. Maximum movement in radial direction is obtained for

Aluminium, Copper and Silver contaminants on the conical spacer inclined

at 450 with respect to enclosure and results are obtained. The influence of
 432

particle location on the breakdown is thus compared by considering the

particle on the enclosure as well as on the spacer for various positions on

the spacer given in fig 3.12. The contaminant on the conical spacer nearer

to the hV conductor attains maximum radial movement as the electric field

experienced by the particle near to the conductor is more. (For Example a

Aluminium contaminant at different positions in fig 3.12 attains maximum

radial movement with application of 200kV power frequency voltage is

11.48874mm at position 1, 22.13959mm at position 2 and 33.16336mm at

position 3 at heights of 10mm, 20mm and 30mm from the enclosure

surface. Thus maximum movement achieved at positions 1, 2 and 3 is

1.48874 mm, 2.13959mm and 3.16336mm respectively with respect to their

initial position. The maximum movement for the same contaminant on the

enclosure under identical condition is 61.1385mm). The maximum

movement in radial direction for the contaminant at position 3 near to hV

conductor is maximum. However, the movement of the contaminant on the

spacer is less when compared to the movement of the particle on the

enclosure. Thus the presence of contaminant will thus weaken the epoxy

insulation of the spacer initiating the breakdown. Thus breakdown voltage

at position 3 will be less when compared positions 1 and 2 given in Fig 3.12.

The work reported in this thesis also deals with the movement of the

metallic particles in a three phase common enclosure busduct with bare

electrodes. Wire type and spherical type contaminants have been

considered. Simulation is carried out for the presence of contaminant in

500/64 GIB. 0.1mm radius and 10mm length wire type particle and

spherical type particle of 0.3mm and 0.4mm radius are considered. Further

these investigations are carried out under pure SF6 as well as with SF6 + N2

gas mixtures.

The variation in maximum radial movement with variation in %

concentration of N2 gas mixtures is presented. Movement of particle for 0%

concentration of N2 in gas mixtures corresponds to pure SF6 and Movement

of particle for 100% concentration of N2 in gas mixtures corresponds to pure

N2. Even with mixture of gases the copper particle for 300kV and Silver

particle for 400kV could not leave the surface. The maximum radial

movement for Aluminium particle with applied power frequency voltage of

600kV for 40%,50% and 60% of N2 in SF6+N2 gas mixtures is 43.5410mm,

43.6828mm and 43.2851mm against 44.5626mm in the case of pure SF6 .

For Copper particle with applied power frequency voltage of 600kV for

40%,50% and 60% of N2 in SF6+N2 gas mixtures was found to be

17.3297mm, 21.0562mm and 19.8878mm when compared to pure SF6 case

wherein the maximum radial movement being 22.0731mm .

The effect of variation of parameters like radii and length of particles,

co-efficient of restitution and pressure on the movement of metallic

contaminant in the busduct also the effect of applied voltage and inner

conductor radii has been presented. It is observed that as the radius of the

particle increases the maximum movement of particle of any material

decreases. With increase in applied voltage the movement of particle

increases. The increase in length of the particle results in the increase in

the movement of the particle. The change in pressure on the movement is

marginal in the case of Copper and Silver particles and there is no impact of

change in pressure on the movement of Aluminium particle.

SCOPE FOR FURTHER STUDY

The present thesis work is carried out for single-phase isolated conductor

busduct and three phase common enclosure busduct for which particle

movement has been simulated under pure SF6 gas as well as with SF6+N2

gas mixtures

The work can be extended to further studies as mentioned below:

The motion of the particle may be assumed to be linear as well as

rotational. 3-dimensional motion can be determined through this

assumption.

Only wire type and spherical contaminants have been considered for

the study, other shapes of the particle may be considered for the

analysis.

The analysis of the particle motion may be done when the inner

conductor is covered with dielectric coatings.

Movement of metallic particle is obtained only under application of

power frequency voltages and further investigations may be carried

out under application of lightning impulse voltage and switching

impulse voltages superimposed on power frequency voltages.

In this thesis the movement of particle at different position one at a

time is considered. But contaminants at different positions at a time

may be considered under application of various types of voltages and

also with and without gas mixtures.

Experimental investigation on movement of various particles in typical

busduct can be carried out.

The actual movement can be recorded using high-speed camera

introducing transparent ports at selected locations.