基于改进灰色模型GM(1,1)的故障间隔时间预测

Prediction of interval time between failures based on improved GM( 1,1)

针对设备故障预测模型难以精确建立的特点,为提高故障间隔时间预测的精确度,提出了变周期三角函数-灰色模型GM(1,1)的预测方法。该方法在三角函数-灰色模型GM(1,1)基础上,建立了变周期三角函数—灰色GM(1,1)的组合模型,实现了对设备故障间隔时间的预测;并将预测结果与三角函数-灰色模型GM(1,1)进行对比,结果表明,采用变周期三角函数-灰色模型GM(1,1)对故障间隔时间进行预测,其预测结果的相对误差由24.16%降到3.24%,提高了预测结果的精度。

It is difficult to accurately establish the model of equipment failure prediction, so the trigonometric function with variable periods--GM （ 1,1 ） was proposed as a method to improve the accu- racy of failure interval time prediction. Based on the trigonometric function-GM（ 1, 1 ）, a combined model of trigonometric function with variable periods-- GM （ 1,1 ） was set up to predict the interval time between equipment failures. The prediction results were compared with trigonometric function- GM （ 1, 1 ）, and results show that by using the new model to predict the time between failures, the relative error of prediction results decreased from 24.16% to 3.24%. Therefore, the new model has greatly improved the prediction accuracy.