This thesis proposes an efficient power system simulator to estimate the automatic sequence of events that follow a fault contingency leading to islanding and cascading outages. The simulator is based on a quasi-steady state model that includes island identification, under-frequency load shedding, over-frequency generator tripping, and island load flow. Contingencies can include the outage of generators, loads, or transmission lines. Often times, a fault of one or two of these power system elements can lead to many cascaded outages and system islanding. The simulator utilizes an innovative method that analyzes the null space of the DC load flow susceptance matrix to identify system islands after each disturbance. Once system islands have been determined, each island power imbalance is calculated and the simulator determines based on the power imbalance in each island whether any load shedding, generator tripping, or primary frequency regulation is required. Once these corrective actions are completed each island will either have been found to balance power or will experience blackout. In the islands that have balanced power, a load flow is computed to see if all line flow constraints are satisfied. Any lines with flow constraint violations are faulted, and the iterative process is repeated under all line flow constraints are satisfied. / The results demonstrate the ability of the simulator to quickly and efficiently predict a system's response to contingencies leading to cascading outages and islanding. Simulations were conducted on a 10-bus 13-line network, a 24-bus 38-line network, and a 72-bus 119-line network. / This thesis also examined the highly complex mixed-integer linear problem of identifying the optimum initial outage in the sense that it would cause the maximum amount of load shedding through islanding. The results on a three-line, three-bus test properly identified the line whose initial outage caused overflows leading to system separation and maximum loss of load.

Fires generate heat and propel burning material into the air above and around the core of the fire. Fires under power lines reduce the breakdown strength of the air insulation due to the influence the heat and particles have on the electric field surrounding the conductors. The result can be flashovers and undesirable power supply interruptions in the electrical transmission network with a considerable impact on Eskom's 275 kV and 400 kV MTS (main transmission system) networks. Eskom typically experiences a loss in sales, a reduction in the quality of the power supplied to consumers and disgruntled consumers who in turn experience financial losses due to a loss in production. In this thesis, the high frequency characteristics of corona and electrical discharges generated by the fire phenomena are studied. The influence of the operating voltages on the electric fields, the potential of different media to initiate ionisation and comparison of conductor construction (bundling and diameter) are all considered in the measurement of high frequency signals in the range of frequencies available in the tuned circuits connected to power lines. The propagation of these high frequency signals is studied both in isolation of other sources of high frequency signals (within controlled laboratory conditions) and in the real environment adjacent to all other interfering sources. Finally the fingerprinting of the varying high frequency signal patterns associated with fires is considered with a view to implementing an operational early detection device. Early detection of a fire allows the utility to understand the source of a system fault, manage it effectively and if possible pre-empt possible failure by means of appropriately applied standard operating guidelines (SOG) to minimise the impact. / Thesis (M.Sc.Eng.)-University of Natal, Durban, 2003.

No description available.

No description available.

No description available.

No description available.

No description available.

Today's societies are critically dependent on the reliable supply of electricity. The reliable operation of large synchronous generators in power stations is therefore of paramount importance. Thus, power utilities are most keenly concerned in taking measures to avoid catastrophic failures and to minimise the impact of generator outages due to faults. Each generator constitutes one of the most expensive single pieces of equipment in the power station. Although they are mostly reliable and generally require minimal maintenance, faults can and do occur with varying degrees of consequences ranging from minor outages to catastrophic failures. / When faults occur, cause-effect relationships need to be identified systematically. This is necessary for having a better understanding of the fault formation mechanisms so that preventive measures can be taken, although sometimes the immediate pressures for returning a generator to service may impede a thorough investigation. / This thesis is borne of the Author's experiences in dealing with large synchronous generators of various sizes and designs. The purpose of this thesis is to report on practical problems in the operation of large synchronous generators and offer advice on remedy. It presents carefully documented cases of faults. These vary from fractured stator conductors and core overheating to cracking of rotor retaining rings and winding short circuits. Symptoms of various faults are discussed along with the extent of damage and remedial solutions. Emphasis is placed on being able to prevent the fault recurrence and prolong the mean time between failures (MTBF). / Thesis (MEngineering(Research))--University of South Australia, 2006.

Thesis (MEng) -- University of South Australia, 1991

Thesis (MEng) -- University of South Australia, 1991