| The nature of the bulk power system reliability problem requires the application of special techniques to handle the complex relationships between the system structure, the utility operating policies and the random behaviour of system variables such as the generation dispatches, load demands and network configurations. As a result of random inputs the performance of the system, the voltage magnitude and angles, active and reactive power flows, is characterized by random behaviour. Furthermore all the reliability measures depend on the assumption concerning operator action in response to changes in system behaviour.; In this study, several techniques to account for the random behaviour of system variables related to real power flow computations, and to represent various operating policies are developed or modified. Modeling of operator action to balance the system is introduced in the context of probabilistic load flow analysis. Maintaining power flows within transmission capabilities is discussed in relation with reliability evaluation of a flow network with random arc capacities using Boolean algebra techniques. In addition, operator action to minimize the amount of load curtailed due to generation and transmission capacity limitations is studied in the case of discrete probability distributions describing system variables.; Some of the most difficult problems in bulk power system reliability evaluation are related to enormous computational effort required in many of these studies. These requirements are the consequences, partly of the large system size and the resulting large numbers of various system conditions that must be assessed, and partly, of the complexities of the failure effects analysis. In this study, the problem of reducing computer running times is solved by developing an implicit enumeration scheme to reduce the number of system conditions tested, and by introducing extremely fast Interval Programming algorithms which exploit the special structure of power system problems.; An efficient algorithm to compute reliability measures given the set of minimal cuts has also been developed and implemented in the flow network reliability evaluation program.; Several numerical examples are produced to illustrate new concepts and techniques. In the concluding chapter, a large bulk power system reliability problem is solved in which many of the above techniques are applied. The computational efficiency of the proposed approach is discussed. |
