摘要
预测船舶升沉运动有助于增强波浪补偿系统的补偿效果,解决补偿系统滞后问题。为提高预测模型的预测精度,提出一种基于误差反向传播(BP)神经网络和长短时记忆(LSTM)神经网络组合优化的船舶升沉运动预测方法。以采用计算流体动力学(CFD)方法获取的船舶在规则波浪作用下的升沉运动和在突发性干扰下的升沉运动为对象,基于PYTORCH框架和LINGO软件,建立以加权方式组合优化BP神经网络和LSTM神经网络的预测模型。研究结果表明,无论是船舶在规则波浪作用下的升沉运动,还是船舶在突发性干扰下的升沉运动,BP-LSTM组合模型的预测精度均高于BP神经网络和LSTM神经网络,有助于提高补偿精度。
The prediction of vessel’s heave motion is helpful to enhance the compensation effect of wave compensation system and solve the lag problem of compensation system. In order to improve the accuracy of prediction model, a vessel’s heave motion prediction method based on the combination optimization of the error back propagation(BP) neural network and the long and short time memory(LSTM) neural network is proposed.The vessel’s heave motion under regular waves and sudden disturbances which are obtained by computational fluid dynamics(CFD), are taken as objects. Based on the PYTORCH framework and the LINGO software, a prediction model is established to optimize the BP neural network and the LSTM neural network in a weighted way. The results show that the prediction accuracy of BP-LSTM combined model is higher than that of the BP neural network and the LSTM neural network for the vessel’s heave motion under regular waves and sudden disturbances, which is helpful to improve the accuracy of compensation.
作者
董冠男
许媛媛
李广健
赵希旺
DONG Guannan;XU Yuanyuan;LI Guangjian;ZHAO Xiwang(Logistics Engineering College,Shanghai Maritime University,Shanghai 201306,China)
出处
《船舶工程》
CSCD
北大核心
2022年第3期55-60,163,共7页
Ship Engineering
基金
国家自然科学基金(61603246)。