Swing Process Model Design of a Cutter Suction Dredger Based on RBF-ARX Model

3rd International Conference on Automation, Control and Robotics Engineering

工业技术;计算机科学;无线电电子学

Jiang, Shuang^1 ; Ni, Fusheng^1 ; Qian, Lina^1

Engineering Research Center of the Education, Ministry of Dredging Technology, Hohai University, 200 jinling north road, changzhou city, jiangsu province, China^1

关键词: Cutter suction dredger;    Dynamic characteristics;    Formation process;    Hidden layers;    Model outputs;    Non-linear parameters;    Process Modeling;    Slow convergences;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/428/1/012018/pdf
DOI  :  10.1088/1757-899X/428/1/012018

来源: IOP

【 摘 要 】

The swing process of a cutter suction dredger are affected by many factors, which changed time and nonlinear. So it is difficult to describe the change law accurately. The swing process model based on RBF-ARX model is established after detailed analysis of the formation process of dredger production. The shortcomings of the traditional neural network could be overcome by this method, such as slow convergence rate, more hidden layer. After modeling, the linear and nonlinear parameters of the RBF-ARX model are identified off-line by using SNPOM (Structured Nonlinear Parameter Optimization Method). Then the errors of the output of the model and the real data are compared and analyzed by simulation. The result show that the swing process model can accurately describe the dynamic characteristics of the system in the global range, and the model output is well fitted with the real data of a cutter suction dredger.