为实现复杂系统的视情维修,对故障预测的需求越来越高。将信息粒化概念引入支持向量机,提出基于模糊信息粒化支持向量机的预测算法,在分析三级无刷交流发电机脉宽调制调压器故障特征的基础上,实现对导通率的区间预测。实验结果表明,预测算法的准确率较高且可行有效,具有实际应用价值。 In order to realize the situation-based maintenance of complex systems, the demand for fault prediction is getting higher and higher. The concept of information granulation is introduced into SVM, and a prediction algorithm based on fuzzy information granulation support vector machine is proposed. Based on the analysis of fault characteristics of pulse width modulation regulator of three-stage brushless alternator, Interval prediction. Experimental results show that the accuracy of the prediction algorithm is high and feasible and effective, which has practical value.