Composite system reliability evaluation using state space pruning
Academic Article
Overview
Additional Document Info
View All
Overview
abstract
Composite system reliability evaluation is concerned with the determination of reliability indices of a power system considering both generation and transmission line capacities and outages. This topic should stay important even in the deregulated environment although the emphasis may be more on load point indices rather than the system indices. Given generation and transmission capacities at various buses, the reliability of major load points would be of interest. Many papers have been written on this subject proposing various methods. These methods can be broadly classified as based on contingency enumeration, state space decomposition and Monte Carlo simulation. Contingency enumeration is theoretically the most accurate approach but it is impossible to apply it directly as the number of contingencies can explode exponentially. Many techniques have been proposed to circumvent this curse of dimensionality. These techniques include state merging, state space truncation and contingency ranking. State space decomposition tries to solve the problem by grouping states into sets having similar properties. This procedure depends on the coherency of state space that can be ensured only if capacity flow model is used. If AC or DC flow models are used coherency holds only for generation changes but not for those resulting from transmission outages. An other drawback of this approach is that for a large network, the number of sets generated can become very large. Monte Carlo methods can use any flow model but their computation time can become long especially for highly reliable systems. Variance reduction techniques have been proposed in the literature as means for reducing the computation time. This paper presents a method for composite system reliability evaluation by performing Monte Carlo simulation selectively on those regions of state space where loss of load states are more likely to occur. These regions are isolated by pruning the state space to remove coher-ent acceptable subspaces. Monte Carlo simulation is then performed on the residual conditional state space. A technique is also introduced for circumventing the problem of noncoherency during pruning. This is achieved by keeping the transmission lines at their maximum capacity during pruning, but allowing them to fail during sampling. Probabilities of generation failures are much higher than those of transmission lines, and this allows a large portion of probability space to be pruned The method assumes a DC flow model and is tested using the modified IEEE Reliability Test System. It is shown that this method results in significant reduction in the number of sampled states, thereby reducing the computational effort required to compute the system and bus indices. The proposed method is not intended to replace existing variance reduction techniques; in fact, such techniques may be used in conjunction with the proposed method to further improve its efficiency.
