A fuzzy-based approach for generation system reliability evaluation

A new method for calculating generation system reliability using fuzzy mathematics is presented here. In reliability evaluation of power systems, a membership function can be provided to represent the uncertainties and inaccuracies in the failure rates and repair rates of generating units. The basic unit addition recursive algorithm is modified to represent these uncertainties along with uncertainties in system load as fuzzy data. Loads are represented as clusters to overcome the difficulty of large computational time requirements. A computer program has been developed in C language to include all aforementioned factors. The program was tested with the IEEE Reliability Test System on an IBM RS/6000 workstation. Several cases were analyzed and results are presented. The algorithm developed computes reliability indices such as loss of load probability (LOLP) and the expected unserved energy (EWE) which are viewed as possibility distributions. (C) 2000 Published by Elsevier Science S.A. All rights reserved.