基于信息融合和改进UGM(1,1)模型的故障预测

针对故障预测中存在数据采样时间间隔不均匀、采样难度大、数据量小等问题，借鉴信息融合技术和灰色预测理论，提出了一种基于信息融合和改进不等时距灰色模型(improved unequal interval grey model， IUGM)的预测方法。首先，运用一元二次非线性回归思想建立不等时距灰色预测模型，并通过初始值改进、残差修正和新陈代谢相结合的方式对模型进行改进；然后基于加权思想提出隶属度加权法，以确定特定个体和同类产品的隶属度权值；最后基于加权思想和IUGM(1,1)模型建立特定个体的故障预测模型。实例仿真验证了所提方法的有效性。

Fault prediction based on information fusion and improved UGM(1,1) model

In view of the problems of unequal date sampling time interval, the great difficulty of date sampling and the small date capacity in fault prediction, a prediction method based on information fusion and improved unequal interval grey model (IUGM) according to information fusion technical and grey prediction theory is poposed. Firstly, unequal interval grey model is built by nonlinear regression idea and optimized by initial value improved, residual error correction and metabolism. Then the degree of membership weighted method is put forward based on weighted thought so as to determine degree of membership value of models. Finally, the fault prediction model of special individuality is built on the basis of degree of membership value and IUGM(1,1) model. The results of simulating example validate the validity of the proposed method.