基于遗传神经网络的耙吸挖泥船泥泵转速预测

Pump Speed Prediction for Hopper Dredger Based on Genetic Neural Network

摘要:

耙吸挖泥船泥泵管线模型是一个复杂的、非线性的动态模型,影响模型准确性的参数较多;为了根据当前施工条件和流量的优化值准确地预测转速,为施工人员提供参考,提高疏浚效率,采用了遗传算法改进的BP神经网络对泥泵转速进行预测;首先,遗传算法对BP神经网络的初始权值和阈值进行优化;然后,BP神经网络根据优化值对网络进行训练并对转速进行预测;为了验证该方法的有效性,将遗传BP神经网络的预测输出和实测泥泵转速进行对比;仿真结果表明:遗传BP神经网络具有很强的非线性拟合能力和全局搜索能力,能够准确地预测泥泵转速;该预测输出可为施工人员提供参考,以便改变泥泵转速,提高疏浚效率。

Abstract:

Hopper dredger's pump pipeline model is a complex and nonlinear dynamic model, and there are lots of parameters that can affect the model's accuracy. In order to accurately predict the next moment's pump speed and improve the dredging efficiency based on current construction conditions and the optimal flow rate, the genetic BP neural network prediction model is proposed. First, genetic algorithm was used to optimize the initial weights and thresholds of BP neural network, and then the BP neural network is trained according to the optimal value. In order to verify the validity of the method, the genetic BP neural network and the real pump data were compared. The simulation results show that the genetic BP neural network has a good fitting ability and good global search ability. Genetic BP neural network can accurately predict the speed and provide recommendations for the construction personnel, who can adjust pump speed and improve the efficiency of dredging.