Weibull Parameter Estimation Algorithm Based on Ant Colony

The Weibull distribution has high fitting accuracy and applicability to the life distribution of rolling bearings. Based on the assumption that life of rolling bearings obeys the Weibull distribution, the Weibull distribution parameters are estimated by using least-squares method (LSM), steepest descent method (SDA) and maximum likelihood estimation (M-L), meanwhile, the ant colony algorithm (ACO) is introduced to determine 2 parameters in Weibull distribution. The comparison and analysis of the fitting accuracy and operational efficiency of 4 kinds of algorithms under the index of fitness and running time are given. The results show that the applicability of ACO algorithm in estimating extremum parameters is proved.