考虑时滞影响的绞吸挖泥船产量预测

Production Prediction of Cutter Suction Dredger in Consideration of Time Delay

为了提高绞吸挖泥船产量的预测精度,提出带有时滞的BP神经网络预测模型。重点分析泥浆浓度与真空度之间的时间延迟和滞后问题,以绞刀电机电流、流速、真空度、横移速度作为BP神经网络模型的输入变量,以泥浆浓度作为输出变量,采用改进后的BP神经网络建立绞吸挖泥船产量预测模型。实验结果表明,该方法能够很好地预测绞吸挖泥船产量,可用于实际挖泥船产量的预测与评估。

The prediction model of BP neural network with time delay is proposed in order to improve the prediction accuracyof cutter suction dredger(CSD). The problem of time delay and lag between slurry density and the degree of vacuum is intensively an-alyzed. The proposed model is built on the basis of the input factors of BP neural network model,i.e.,the current of the cutter head,the velocity of pipeline transportation,the degree of vacuum and the swing speed,and the output factor,i.e.,slurry density. The re-sults show that the method can well predict the dredger production and can be used to predict and evaluate the actual production of aCSD.