As a key part of various mechanical equipments,rolling bearing often affects the running condition of whole machine equipment.Therefore,highprecision rolling bearing state trend prediction has important significance for the running state of whole machine equipment.A combined forecasting model based on time series and support vector regression machine theory was constructed for the problem of poor accuracy of rolling bearing single prediction model.Firstly,single model prediction was done,the error squared and reciprocal method was used to obtain the weighting results of the two prediction models.Finally,the prediction results of the combined model were compared with those of the single prediction models.The experimental results show that the combined forecasting model has less prediction error than the single model and has higher reliability.