Avaliação da confiabilidade preventiva de sistemas de potência

Abstract

The regulatory reform of the electric power industry creates an entirely new competitive environment. In this new electricity market, the reliability of services plays a very important role to establish non-deterministic criteria, to be applied to both operation and expansion planning of electric power systems. The utilization of these criteria, however, is being slowly incorporated into the decision-making processes of most utilities. Due to the difficulties of interpreting numerical risk indices, system operators and planners are still averse to the use of probabilistic techniques, being more confident with the traditional deterministic criteria. The well-being analysis has been recently proposed as a new technique to measure the degree of adequacy of power systems, which incorporates deterministic criteria in a probabilistic framework. The combination of the basic deterministic and probabilistic concepts is established through the classification of the system operating states into three categories: healthy, marginal and at risk. In order to identify these states, the system is analyzed according to a deterministic criterion based on a pre-specified list of equipment contingencies. In this dissertation, a new methodology is proposed to evaluate well-being indices considering composite generation and transmission power systems. The new methodology uses a non-sequential Monte Carlo simulation, a non-aggregate Markovian load model and a new process to estimate failure frequency indices, named one step forward state transition process. New test functions are proposed to calculate the well-being indices, including the frequency of marginal states. In order to test the accuracy and efficiency of the proposed methodology, the IEEE Reliability Test System with some modifications is used.