摘要
提出了基于EKF神经网络的疏浚作业过程动态演化建模方法。在神经网络建模过程中引入卡尔曼滤波思想,利用扩展卡尔曼滤波实时更新神经网络模型的权重,从而获得能有效跟踪挖泥船疏浚过程工况变化的模型。
This paper presents a areaglng dy- namic evolution modeling method based on an ex- tended Kalman filter neural network algorithm. The concept of the Kalman filter is introduced in the process of neural network modeling. Using the extended Kalman filter real - time updating weights of the neural network model, effective models showing the variation of dredger dredging process conditions can be tracked.
出处
《机械与电子》
2015年第6期23-25,29,共4页
Machinery & Electronics
基金
中央高校基本科研业务费专项资金资助(2014B10122)
关键词
疏浚作业
扩展卡尔曼神经网络
能耗与产量
模型
dredging
extended Kalman filterneural network
energy consumption and produc-tion
model
