柴油机故障诊断中的变权重多证据体融合应用研究

为了能够确定各证据的权重,对证据进行权重调整,应用了基于遗传算法的加权平衡证据调整方法。首先,利用遗传算法确定最优化的或者近似最优化的权重系数,然后依据加权平均值和优先权的证据分布形态对证据进行加权调整,且保证调整后的证据加权平均值不变以及优先权证据分布形态稳定。柴油机故障诊断实例表明,采用基于遗传算法的加权平衡证据调整方法能提高故障诊断的信任度和准确率,明显优于普通D-S等权重证据组合方法。

Application Research on Weight-varying Multi-evidence Fusion in Fault Diagnosis of Diesel Engine

To determine the weight of each evidence,the method of weighted evidence balance based on the genetic algorithm was used to adjust the weight of evidence.The optimal or nearly optimal weight coefficients were determined by the genetic algorithm and then the evidences were weighted and adjusted according to the weighted average and distribution of preferred evidence that should be constant before and after the balance.The experimental results show the weighted evidence balance method can improve the confidence and accuracy of fault diagnosis,which outgoes the basic evidence theory method.