Reliability Prediction Methods and Application of Large Capacity Generating Units Available to Purchase

The mathematical model and the methodology of the reliability prediction of generating units are presented. Based on statistical analysis of operation reliability past data for generating units, statistical values of the repair factor and the mathematical model’s parameters of the repair factor are determined. According to plan repair outage days and the mathematical model for the repair factor of some generating unit, equivalent availability factor (EAF) of the generating unit can be predicted in future three years. The reliability prediction examples for sub-critical 300MW, supercritical 600MW and sub-critical 600MW fossil units are given together with reliability prediction results of 550MW hydro units and 984MW, 990MW nuclear units. The relative error’s range for equivalent availability factor prediction values of the generating units is between −1.48% and 2.69% which indicates that reliability prediction precision is higher. By using reliability prediction method, prediction values for the reliability indexes of generating units can be quantitatively calculated, which provides a basis for reliability objective management and optimization repair of generating units.