Thesis (M.Sc.) - Cairo University - Faculty of Engineering - Department of Electrical Power and Machines

One of the most serious problems that cause the failure of the GIS substations is the presence of contaminating particles inside its compartments. This thesis studies the initiation of particle-based discharges in the gas. Simulation using finite-elements techniques is carried out as the first step to enable the determination of electric fields around particles. Ionization volume occurs around surface of these particles. Based on gas properties the boundary of that ionization volume is obtained and the discharge volume is subsequently calculated. An identification factor k is generated by linking the discharge volume to the probability of discharge initiation time and it is used in assessing interaction between particles and the associated shielding effect. Particle presence diagnosis is based in this thesis on the detection of time-to- discharge. A comprehensive database of different scenarios of particles presence is first computationally compiled and then with mean discharge time of distribution times uniquely identifies a particle condition scenario which is obtained from the simulation analysis. After this, matching between this and the actual distribution of the mean discharge times which is detected experimentally from the live laboratory testing on a given GIS apparatus. This matching of the two sets of data {u2013} computational and experimental- is the basis for contaminating particle diagnosis.

Issued also as CD