Development of model for assessment of operational reliability and power distribution reliability of a power generating system

Abstract

In modern day life the requirement of power supply is increasingly becoming an absolute necessity. Efficient and effective power supply hinges on two factors, namely effective power generation and effective distribution. From the engineering view point, operational reliability is considered as a measure of system .effectiveness. Operational reliability, in turn, is not based on a single quantitative measure. The most common premises have been, time-dependent reliability, steady-state availability and the like. In the present work, plant reliability of Khulna steam power station and distribution system reliability of Khulna industrial zone have been studied. Power generation by steam is by far the cheapest but problematic. Any steam power plant consists of a large ,number of complex equipments which are liable to failure due to various causes. From the failure and maintenance records of the said plant, reliabilities of individual equipments and the overall system reliability have been calculated. Based on these, critical equipments have been identified and their modes of failures have been analysed. Possible reliability improvement by eliminating principal mode of failure of each critical equipment has 6een demonstrated. This has again been compared to reliability improvement which may be effected by installing stand~~) unit of critical equipment. Overall system reliability improvement has also been shown. vi Distribution system failures have been analyzed and found to be random in nature. For this case, reliability indices have been considered to be failure rate, mean failure duration and availability. These indices have been calculated for individual substations, feeders and load points. Overall system failure rate and mean failure duration have also been assessed. For time-dependent reliability, Weibull reliability function has been used throughout. Disttibuti'on parameters have been calculated by Maximum Likelihood Estimator and hazard plotting methods.