疏浚系统泥浆浓度的自校正前馈控制

Self-tuning feedforward control of slurry concentration in a dredging system

中文关键词  疏浚作业  浓度控制  前馈控制  自校正控制

英文关键词  dredging operation  concentration control  feedforward control  self-tuning control

中文摘要

      采用CARMA模型表示泥浆浓度过程, 并通过现场的实验数据, 利用最小二乘算法对泥浆浓度过程的数学模型进行了辨识; 在不同工作点上的辨识结果表明泥浆浓度过程具有参数时变和外部干扰作用明显等特点. 本文提出了一种自校正控制和前馈补偿相结合的控制方案; 控制系统利用在线检测的过程数据, 采用递推最小二乘算法连续更新系统前向通道和可测量干扰通道的传递函数. 利用更新的系统模型设计控制器参数, 控制器的调节目标是使系统未来时刻的总体方差趋于最小. 以900型挖泥船为基础, 通过实验的方式检验了本文提出的控制方案的性能. 实验结果表明该方法在系统作业环境变化, 时滞较大的条件下仍能使泥浆浓度基本保持稳定, 具有较强的抗干扰能力和良好的跟踪性能.

英文摘要

      A CARMA model is introduced to represent the slurry concentration process. With experimental data obtained on-site, this model is identified by using the least-squares method. Identifications, which are carried out on different working locations, show that the model parameters of a slurry concentration process varies with time, and external influences exhibit remarkable impacts on the process. A control scheme, incorporated self-tuning control with feedforward control, is introduced for controlling the slurry concentration. With process data acquired online, the control system continuously updates the process model by using the Recursive-Least-Squares (RLS) method. With the updated process model, a control law is determined to minimize the sum-square-error beyond future steps. Experiments have been carried out on a type-900 dredger. Results show that the control performance is satisfactory in various dredging environments, even though the time-lag is significant. The control system effectively compensates external disturbances, and has excellent tracking ability.