【摘要】 结合超声波堰式流量测试方法的研究 ,在信号处理领域首次引进了“滑动窗口”的概念 ,提出了一种滑窗式模型预估的算法 ,给出了RBF神经网络实现模型实时预估的方法 ,解决了液面不平引起流量测试故障实时恢复的问题 ,提高了堰式流量测试的精度。滑窗式模型预估的算法对其他物理量的测试同样有一定的指导作用。更多还原

【Abstract】 In accordance with the research of ultrasonic barrier flow volume measure. the conception of sliding window is applied to the signal measured. The algorithm of model prediction with RBF neural network is described. The fault caused by the fluctuation of the liquid surface is diagnosed and recovered. The precision of barrier flow volume measurement is improved. The algorithm of sliding window model prediction will be helpful to the research of other similar measurements.更多还原