APP下载

基于改进PSO-ARIMA模型的船舶纵摇角度预测

2021-04-30王培良张婷肖英杰

上海海事大学学报 2021年1期
关键词:船舶模型

王培良 张婷 肖英杰

摘要:

针对自回归移动平均(auto regressive moving average, ARMA)模型在船舶纵摇角度预测时不具有普遍适用性问题,提出使用自回归综合移动平均(auto regressive integrated moving average, ARIMA)模型进行纵摇角度预测,并采用改进粒子群优化(particle swarm optimization, PSO)算法对模型定阶。对纵摇角度值序列数据进行平稳性检验和差分运算,确定ARIMA模型的适用性;采用具有针对性适应度评价函数的PSO算法进行模型定阶,并优化PSO算法的权重计算方法。通过仿真对比验证本文所提方法的科学性和有效性。仿真结果表明:采用改进PSO算法进行模型定阶的方法能够有效提升模型的预测精度,具有更好的预测效果。

关键词:

自回归综合移动平均(ARIMA)模型; 粒子群优化(PSO)算法; 船舶纵摇; 纵摇预测

中图分类号:  U661.32+1

文獻标志码:  A

收稿日期: 2020-07-15

修回日期: 2020-09-22

基金项目: 国家自然科学基金(51909155);潍坊市科学技术发展计划(2019GX075)

作者简介:

王培良(1987—),男,山东潍坊人,博士研究生,研究方向为载运工具运用工程,(E-mail)gfy5216@126.com;

张婷(1987—),女,山东聊城人,讲师,硕士,研究方向为航海技术,(E-mail)titi-507@163.com;

肖英杰(1959—),男,广东潮阳人,教授,博导,船长,研究方向为航海技术,(E-mail)xiaoyj@shmtu.edu.cn

Prediction of ship pitch angle based on improved PSO-ARIMA model

WANG Peiliang1,2a,2b, ZHANG Ting3, XIAO Yingjie2a,2b

(1.School of Intelligent Manufacturing, Weifang University of Science and Technology, Weifang 262700, Shandong, China;

2.a.Merchant Marine College; b.Engineering Research Center of Shipping Simulation,

Ministry of Education, Shanghai Maritime University, Shanghai 201306, China;

3.Navigation College, Shandong Transport Vocational College, Weifang 261206, Shandong, China)

Abstract:

In view of the fact that the auto regressive moving average (ARMA) model is not of the universal applicability when predicting the ship pitch angle, an auto regressive integrated moving average (ARIMA) model is proposed for predicting the ship pitch angle, and an improved particle swarm optimization (PSO) algorithm is adopted to determine the model order. For the pitch angle value series, the stationarity test and difference operation of the data are performed to determine the applicability of the ARIMA model. The PSO algorithm with a targeted fitness evaluation function is used to determine the model order, and the weight calculation method of the PSO algorithm is optimized. The scientificity and effectiveness  of the method proposed in this paper is verified through the simulation. The simulation results show that the method using the improved PSO algorithm determining the model order can effectively improve the prediction accuracy of the model and makes the model have better prediction effect.

Key words:

auto regressive integrated moving average (ARIMA) model;particle swarm optimization(PSO) algorithm; ship pitch; pitch prediction

0 引 言

船舶航行时受到外界环境(如风、浪、流等相互作用)的影响,产生六自由度的摇摆运动,严重威胁船舶的航行安全,因此,针对船舶摇摆随时间变化规律的预测研究是航运界的研究难点和热点[1]。

时间序列分析法是将预测对象的相关属性值按照时间顺序排列,然后结合数学模型研究其属性变化规律,从而对属性值进行预测[2-3]。文献[4]通过在护卫船上安……

登录APP查看全文

猜你喜欢

船舶模型
一半模型
计算流体力学在船舶操纵运动仿真中的应用
基于改进谱分析法的船舶疲劳强度直接计算
重尾非线性自回归模型自加权M-估计的渐近分布
船舶!请加速
BOG压缩机在小型LNG船舶上的应用
船舶压载水管理系统
3D打印中的模型分割与打包
FLUKA几何模型到CAD几何模型转换方法初步研究
一个相似模型的应用