Abstract:The life of woven liner self-lubricating spherical plain bearing is mainly determined by the wear performance of the liner,which is generally non-linear,making it difficult to predict the life.Based on the method of simultaneous change of load and swing frequency for accelerated life test,a gray neural network prediction model was established with pv value,wear amount degradation data as input parameters and life value as output parameters.It was verified that the maximum error predicted by the prediction model is only 7.33%,and the average error is only 3.892%.The reliability of selflubricating spherical plain bearing under different acceleration stresses was evaluated.It is concluded that the reliability of the spherical plain bearing is decreased slowly before L10(lifetime at 90% reliability),and then decreased rapidly.The greater the pv value,the faster the reliability decreases.As the pv value increases,the life of the spherical plain bearing is decreased exponentially,the inverse power law model can be used to reflect the relationship between the model responses.