Abstract: During the dredging operation, the main operation parameters of cutting suction dredger are decided by operators, whose personnel experience and theoretical level is comparatively large difference. The operators make?a decision according to their own experience, so the manual operation is far lower than expected yield, lower efficiency, higher operation cost, poorer operation quality. In order to overcome these shortcomings, this paper puts forward a dredging state space modeling method??based on BP neural network, and designs linear quadratic optimal control tracker, which is applied to cutter suction dredger swing process control and is an effective method of the intelligent control instead of the manual operation. The simulated result shows that the linear quadratic optimal tracking system has more stabilization, quicker response, smaller hysteresis, better tracking of the desired output, and overcomes the drawback of manual operation.

Key words: dredger, dredging system, state space, model, quadratic, optimum control, strategy

摘要： 在疏浚作业过程中，绞吸挖泥船主要的作业参数是由操作人员根据自己的经验确定的，而疏浚操作人员的经验和理论水平也相差较大，所以手动作业的实际产量要远远低于预计产量，效率低下，作业成本偏高，作业质量较差。为了克服该缺点，提出一种基于BP（Back-Propagation）神经网络的疏浚横移过程状态空间建模方法，并设计了疏浚横移过程线性二次型泥浆浓度最优控制跟踪器，应用于疏浚作业智能化控制中。仿真结果表明：二次型最优控制跟踪器的控制稳定性好、响应快、滞后小，跟踪效果好，泥浆浓度提高显著，较好地克服了人工操作缺点。

关键词: 挖泥船, 疏浚系统, 状态空间, 模型, 二次型, 最优控制, 策略