论文部分内容阅读
为了提高列车轮对故障诊断准确率和改善现有列车轮对状态在线监测方法的不确定性,结合多传感器信息融合原理,设计了列车轮对融合监测系统,采用特征层融合自适应加权算法进行了轮对状态融合监测,以自适应的方式寻求最优加权因子,使状态测量值总均方误差最小,比较了特征层融合自适应加权算法、模糊数据关联算法、变结构多模的状态估计算法和BP神经网络算法的计算结果。比较结果表明:当轮对两端轴承均出现故障后,两传感器输出的测量值分别为22.0470和21.0250,而此融合算法计算出的估计值为4.2642,融合值最接近真值,因此,列车轮对融合监测系统可靠性高,抗干扰性强。
In order to improve the diagnostic accuracy of train wheelsets and to improve the uncertainty of on-line monitoring methods of existing train wheelsets, a train wheelset fusion monitoring system is designed based on the principle of multi-sensor information fusion. The feature-level fusion adaptive weighted algorithm In this paper, the wheel-to-wheel state fusion monitoring is adopted to find the optimal weighting factors in an adaptive manner, which minimizes the total mean square error of the state measurements. The feature-level fusion adaptive weighting algorithm, fuzzy data association algorithm and state- Algorithm and BP neural network algorithm results. The comparison results show that the measured values of the two sensors output are 22.0470 and 21.0250 respectively when the bearings on both ends of the wheel are out of order, and the estimated value calculated by the fusion algorithm is 4.2642, the fusion value is the closest to the true value. Therefore, The fusion monitoring system has high reliability and strong anti-interference.