【摘要】 为了描述挖泥船疏浚过程各状态变量在相邻时刻的状态转移变化规律,采用BP神经网络算法对疏浚过程样本数据进行训练分析,以绞吸挖船疏浚作业时的绞刀横移速度作为控制输入,选择对疏浚产量影响显著的两个疏浚过程参数即管道流速和泥浆浓度作为状态变量,同时,选择泥浆浓度作为输出变量,构建了疏浚系统的状态空间模型。通过实际数据与Matlab的仿真数据的比较,所建立的状态空间模型能够较准确地描述疏浚控制系统的动态过程。更多还原

【Abstract】 In order to describe the relationship among variables in dredging system,in this paper,a BP neural network algorithm was adopted for the analysis of the sample data collected during dredging process.The swing speed was employed as the control input,the two process parameters of velocity for pipes and mud concentration were used as state variables,which influenced the dredging output significantly.At the same time,the mud concentration was employed as the output variable,so that the state space model was constructed.The comparison between the actual data and the simulation data of Matlab shows that the established state space model,which is an essential work for the optimal control target of dredging process,can relatively accurately describe the dynamic process of the dredge control system.更多还原