(1997) Preston, Eugene Gordon, 1947-; Grady, W. M.; Baughman, Martin L.

Show more

A new probabilistic load flow (PLF) model for calculating the reliability of large nonequivalenced electric networks with transmission constraints is given. Generation loss of load probability (LOLP) and expected unserved energy (EUE) is calculated first without transmission constraints as a function of load level. Then a two step process is used to 1) calculate the cumulative probabilistic line flows from random generator failures and 2) perform load-generator reductions to remove line overloads. The additional EUE and LOLP due to transmission constraints is calculated. New piecewise-quadratic (PQ) convolution methods are used to accurately calculate probabilistic line flows for the total set of generator failure configurations on every transmission line (>2³⁰⁰ [almost equal to] 10⁹⁰ for the 300 generator Texas system) in a reasonable amount of computation time. Complete coverage of all generator outage configurations resolves problems associated with Monte Carlo and other enumeration methods. A new method for outaging multiple transmission lines allows the majority of probability space of all transmission line outage events to also be calculated in conjunction with the generation outages. A large network example is presented in which the benefit of an additional autotransformer in a large system is calculated. Another example using the IEEE RTS benchmarks the PLF model against a full configuration enumeration with linear programming solution.