为了提高船舶在风浪中航行的安全性,需要精确地预测船舶在风浪中的横摇运动,以提高船舶横摇控制效果,本文通过应用一种带外源输入的非线性自回归(NARX)神经网络预测方法预测船舶横摇运动。该方法考虑了实船操纵性试验数据受风、浪、流等外界因素的的影响,将实测的风向、风速、流速、流向、浪向以及浪高的数据作为外源输入,能够有效提高船舶横摇运动的预测精度。基于“育鲲”轮,利用该方法对实际船舶海上横摇运动进行了实时预测实验,并将其实验结果与SAPSO-BP神经网络模型的预测结果进行对比。从对比结果可以看出,本文所提方法对复杂海浪环境具有良好的适应性,NARX模型的预测精度优于普通反向传播(BP)神经网络和自适应粒子群算法优化的普通反向传播(SAPSO-BP)神经网络.
In order to accurately and efficiently predict the rolling motion of a ship in waves and ensure the safety of personnel, cargo and ship, a nonlinear autoRegressive xogenous (NARX) neural network prediction method based on external input is proposed in this paper. The influence of wind on the actual ship maneuverability test data is considered in this method, and the measured wind direction and wind speed data are taken as external input, which can effectively improve the performance of the ship. The prediction accuracy of ship rolling motion. Based on the scientific research practice ship Yukun of Dalian Maritime University, the real-time prediction experiment of ship rolling motion is carried out by using this method, and the experimental results are compared with the prediction results of simulated annealing particle swarm optimization-back propagation (SAPSO-BP) neural network model optimized by adaptive particle swarm optimization. From the comparison results, it can be seen that the proposed method has good adaptability to complex sea wave environment, and the prediction accuracy of NARX model is better than that of BP neural network and SAPSO-BP neural network optimized by adaptive particle swarm optimization.
2022,44(11): 63-67 收稿日期:2021-06-29
DOI:10.3404/j.issn.1672-7649.2022.11.013
分类号:TP183
基金项目:国家重点研发计划资助项目(2019YFB1600602);辽宁省重点研发计划资助项目(201801704);辽宁省自然基金资金项目(201801705)
作者简介:李冲(1997-),男,硕士研究生,研究方向为交通信息工程及控制
参考文献:
[1] 傅晓斌. 船舶在横浪中的横摇运动及其稳定性研究[J]. 舰船科学技术, 2015, 37(7): 128–131
[2] 蔡韡, 任元洲, 严传续, 等. 基于神经网络的四自由度船舶操纵运动预报[J]. 中国造船, 2013, 54(4): 155–162
[3] FOSSEN T I. . Handbook of marine craft hydrodynamics and motion control[M]. John Wiley & Sons, Ltd: 2011-04-08.
[4] 李晖, 郭晨, 李晓方. 基于小波变换理论的船舶横摇运动分析与预测[J]. 大连海事大学学报, 2010, 36(1): 5–8
[5] 刘胜, 杨震. 船舶横摇运动实时在线预报方法[J]. 电机与控制学报, 2011, 15(10): 82–87+94
[6] 张泽国, 尹建川, 胡江强, 等. 基于灰色模型粒子群优化算法的自适应神经模糊推理系统模型的船舶横摇运动预报分析[J]. 科学技术与工程, 2016, 16(33): 124–129
[7] 朱想, 丁云龙, 郭力, 等. 基于改进NARX神经网络算法的光伏发电功率短期预测[J]. 武汉大学学报(理学版), 2020, 66(5): 505–511
[8] 张隆辉, 刘正锋, 魏纳新, 等. 基于Narx网络模型的系泊缆截断点运动计算[J]. 船舶力学, 2019, 23(8): 936–947
[9] PEDRAM Rgohanc, KAROLY C. K. Quantifying the thermal damping effect in underground vertical shafts using the nonlinear autoregressive with external input (NARX) algorithm[J]. International Journal of Mining Science and Technology, 2018.
[10] 楼梦瑶, 王旭阳, 陈瑞, 等. 基于NARX神经网络的船舶升沉运动实时预测方法[J]. 中国舰船研究, 2020, 15(1): 48–55+67
[11] 张泽国, 尹建川, 胡江强, 等. 基于减聚类ANFIS模型的船舶横摇运动实时预测[J]. 上海海事大学学报, 2017, 38(1): 7–11+67
[12] 李明, 杨汉生. 一种改进的NARX回归神经网络[J]. 电气自动化, 2006, 28(4): 6–11
[13] LEONTARITIS I J, BILLINGS S A. Input-output para-metric models for non-linear systems, part I: determin-istic non-linear systems[J]. International Journal of Control, 1985, 41(2): 303–328
[14] 王红君, 史丽荣, 赵辉, 等. 基于NARX神经网络的日光温室湿度预测模型研究[J]. 广东农业科学, 2014, 41(20): 170–172+177
[15] 陈雯吉. 基于主成分分析和NARX动态神经网络的股市情绪指数构建与预测[D]. 兰州: 兰州大学, 2016.
[16] 张泽国, 尹建川, 柳成. 基于自适应变异PSO-BP算法的船舶横摇运动预测[J]. 舰船科学技术, 2016, 38(23): 69–73