基于CPSO算法改进GM-Markov模型的港口货物吞吐量预测

doi:10.3969/j.issn.1674-0696.2025.08.14

重庆交通大学学报（自然科学版） ›› 2025, Vol. 44 ›› Issue (8): 108-115.DOI: 10.3969/j.issn.1674-0696.2025.08.14

• 交通+大数据人工智能 • 上一篇

基于CPSO算法改进GM-Markov模型的港口货物吞吐量预测

陈丹涌1，王俞亮1，曾枫泓1，吴承禧2

(1.广州航海学院航运学院，广东广州 510725；2. 新加坡国立大学工业系统工程与管理系，新加坡 119077)

收稿日期:2024-10-08 修回日期:2025-02-28 发布日期:2025-09-05
作者简介:陈丹涌（1972—），男，广东潮州人，副教授，硕士，主要从事交通信息工程与控制方面研究。E-mail：1078876425@qq.com 通信作者：吴承禧（2002—），男，广东揭阳人，硕士研究生，主要从事海事技术与管理方面研究。E-mail：wuchengxi@u.nus.edu

Port Cargo Throughput Forecasting Based on CPSO Algorithm Improved GM-Markov Model

CHEN Danyong1, WANG Yuliang1, ZENG Fenghong1, WU Chengxi2

(1. School of Shipping and Maritime Studies, Guangzhou Maritime University, Guangzhou 510725,Guangdong, China; 2. Department of Industrial Systems Engineering and Management, National University of Singapore, Singapore 119077, Singapore)

Received:2024-10-08 Revised:2025-02-28 Published:2025-09-05

摘要/Abstract

摘要： 针对广东揭阳港惠来港区货物吞吐量的非线性动态预测需求，提出一种基于混沌粒子群优化的GM-Markov组合预测模型。通过集成灰色GM(1,1)模型与Markov链的优势，采用Logistic映射实现粒子群参数与状态区间的混沌初始化，构建具有动态适应能力的预测框架；改进后的模型通过状态空间划分与独立概率转移矩阵计算，有效验证了港区2007—2022年吞吐量数据的随机波动特征。研究结果表明：优化模型将平均绝对百分比误差下降至8.06%，较传统方法显著提升了预测精度与稳定性，验证了该模型在动态系统预测中的工程适用性。

关键词: 交通运输工程；灰色马尔可夫理论；混沌粒子群优化算法；惠来港区；货物吞吐量预测

Abstract: To address the nonlinear dynamic prediction requirements for cargo throughput in Huilai Port area, Jieyang Port in Guangdong Province, a GM-Markov combined prediction model based on chaotic particle swarm optimization was proposed. By integrating the advantages of the grey GM (1,1) model and Markov chain, Logistic mapping was employed to achieve chaotic initialization of particle swarm parameters and state intervals, and a prediction framework with dynamic adaptability was established. The improved model effectively verified the random fluctuation characteristics of throughput data in the port area from 2007 to 2022 through state space partitioning and independent probability transition matrix calculation. The research results demonstrate that the optimized model reduces the mean absolute percentage error (MAPE) to 8.06%, which significantly enhances the prediction accuracy and stability, compared to traditional methods. The engineering applicability of the proposed model in dynamic system forecasting is also verified.

Key words: traffic and transportation engineering; grey-Markov theory; chaos particle swarm optimization algorithm; Huilai Port area; cargo throughput forecasting

中图分类号:

U691.71

陈丹涌1，王俞亮1，曾枫泓1，吴承禧2. 基于CPSO算法改进GM-Markov模型的港口货物吞吐量预测[J]. 重庆交通大学学报（自然科学版）, 2025, 44(8): 108-115.

CHEN Danyong1, WANG Yuliang1, ZENG Fenghong1, WU Chengxi2. Port Cargo Throughput Forecasting Based on CPSO Algorithm Improved GM-Markov Model[J]. Journal of Chongqing Jiaotong University(Natural Science), 2025, 44(8): 108-115.

参考文献

［1］孙志林, 卢雅倩, 黄赛花. 港口吞吐量的马氏链-时序分析预测［J］. 浙江大学学报(工学版), 2012, 46(7): 1289-1294.
SUN Zhilin, LU Yaqian, HUANG Saihua. Prediction of port throughput based on Markov chain-time series analysis［J］. Journal of Zhejiang University (Engineering Science), 2012, 46(7): 1289-1294.
［2］封钰, 宋佑斌, 金晟, 等. 基于随机森林算法和粗糙集理论的改进型深度学习短期负荷预测模型［J］. 发电技术, 2023, 44 (6): 889-895.
FENG Yu, SONG Yubin, JIN Sheng, et al. Improved deep learning short-term load forecasting model based on random forest algorithm and rough set theory ［J］.Power Generation Technology, 2023, 44 (6): 889-895.
［3］杨宇鸽, 郝杨杨, 王逸文. 基于Neural Prophet-LSTM组合模型的港口货物吞吐量预测［J］. 中国航海, 2023, 46(4): 85-92.
YANG Yuge, HAO Yangyang, WANG Yiwen. Prediction of port cargo throughput using Neural Prophet-LSTM combination model［J］. Navigation of China, 2023, 46(4): 85-92.
［4］丁天明, 潘宁, 杜柏松, 等. 基于改进灰色马尔可夫的港口货物吞吐量预测研究［J］. 重庆交通大学学报(自然科学版), 2023, 42(9): 130-136.
DING Tianming, PAN Ning, DU Baisong, et al. Forecast of cargo throughput in port based on improved grey Markov［J］. Journal of Chongqing Jiaotong University (Natural Science), 2023, 42(9): 130-136.
［5］刘思峰, 郭天榜. 灰色系统理论及其应用［M］. 2版. 北京: 科学出版社, 1999.
LIU Sifeng, GUO Tianbang. Grey System Theory and Its Application［M］. 2nd ed. Beijing: Science Press, 1999.
［6］杨金宝, 梁勇, 曹现宪. 一种基于灰色理论-隐马尔科夫模型的装备故障预测方法［J］. 舰船电子工程, 2018, 38(8): 128-132.
YANG Jinbao, LIANG Yong, CAO Xianxian. A method of equipment failure prediction based on grey theory-HMM［J］. Ship Electronic Engineering, 2018, 38(8): 128-132.
［7］徐志胜. 安全系统工程［M］. 北京: 机械工业出版社, 2007.
XU Zhisheng.Safety System Engineering［M］. Beijing: China Machine Press, 2007.
［8］ SHAMSHAD A, BAWADI M A, WANHUSSIN W M A, et al. First and second order Markov chain models for synthetic generation of wind speed time series［J］.Energy, 2005, 30(5): 693-708.
［9］杨承莲, 徐厚宝. 基于粒子群优化灰色马尔可夫模型及其应用［J］. 数学的实践与认识, 2023, 53(9): 267-276.
YANG Chenglian, XU Houbao. Grey Markov model optimized by particle swarm algorithm and application［J］. Mathematics in Practice and Theory, 2023, 53(9): 267-276.
［10］马全党, 江福才, 范庆波, 等. PSO-无偏灰色马尔科夫模型在船舶交通流量预测中的应用［J］. 中国航海, 2019, 42(1): 97-103.
MA Quandang, JIANG Fucai, FAN Qingbo, et al. Application of PSO-unbiased grey Markov model in ship traffic flow prediction［J］. Navigation of China, 2019, 42(1): 97-103.
［11］李兵, 蒋慰孙. 混沌优化方法及其应用［J］. 控制理论与应用, 1997, 14(4): 613-615.
LI Bing, JIANG Weisun. Chaos optimization method and its application［J］. Control Theory & Applications, 1997, 14(4): 613-615.
［12］吴月秋, 纪昌明, 王丽萍, 等. 基于混沌粒子群算法的水电站水库优化调度［J］. 人民黄河, 2008, 30(11): 96-97.
WU Yueqiu, JI Changming, WANG Liping, et al. Optimal operation of hydropower stations and reservoirs based on chaotic particle swarm optimization algorithm［J］.Yellow River, 2008, 30(11): 96-97.
［13］高尚, 杨静宇. 混沌粒子群优化算法研究［J］. 模式识别与人工智能, 2006, 19(2): 266-270.
GAO Shang, YANG Jingyu. Research on chaos particle swarm optimization algorithm［J］. Pattern Recognition and Artificial Intelligence, 2006, 19(2): 266-270.
［14］高鹰, 谢胜利. 混沌粒子群优化算法［J］. 计算机科学, 2004, 31(8): 13-15.
GAO Ying, XIE Shengli. Chaos particle swarm optimization algorithm［J］. Computer Science, 2004, 31(8): 13-15.
［15］ CHEN Yakuan, GOLDSMITH J, OGDEN R T. Variable selection in function-on-scalar regression［J］.Stat, 2016, 5(1): 88-101.
［16］ LI Zhijun, WANG Weiwei, CHEN Mianyun. Improved grey-Markov chain algorithm for forecasting［J］.Kybernetes, 2009, 38(3/4): 329-338.
［17］ HU Hongtao, ZHAI Xiaojing, GUAN Xin. Crude oil output forecasting based on PSO of unbiased gray Markov model［C］//2017 8th IEEE International Conference on Software Engineering and Service Science (ICSESS). 2017, Beijing, China. IEEE, 2017: 644-647.
［18］朱红求, 阳春华, 桂卫华, 等. 一种带混沌变异的粒子群优化算法［J］. 计算机科学, 2010, 37(3): 215-217.
ZHU Hongqiu, YANG Chunhua, GUI Weihua, et al. Particle swarm optimization with chaotic mutation［J］.Computer Science, 2010, 37(3): 215-217.
［19］ POLLOCK A C, MACAULAY A, DILEK  A, et al. Evaluating predictive performance of judgmental extrapolations from simulated currency series［J］.European Journal of Operational Research, 1999, 114(2): 281-293.
［20］ AGRAWAL S, SILAKARI S, AGRAWAL J. Adaptive particle swarm optimizer with varying acceleration coefficients for finding the most stable conformer of small molecules［J］.Molecular Informatics, 2015, 34(11/12): 725-735.

[1]	李广儒，张新，朱庆辉. 基于Adaboost的改进Elman神经网络港口吞吐量预测方法[J]. 重庆交通大学学报（自然科学版）, 2021, 40(05): 1-5.
[2]	李广儒，朱庆辉. 基于Elman神经网络的港口货物吞吐量预测[J]. 重庆交通大学学报（自然科学版）, 2020, 39(06): 8-12.
[3]	王凤武，张晓博，吉哲，王乐. 基于多变量LSTM模型的青岛港集装箱吞吐量预测[J]. 重庆交通大学学报（自然科学版）, 2022, 41(10): 54-61.
[4]	邓萍1，刘淑龙2. 基于FGM-SVR组合模型的港口吞吐量预测[J]. 重庆交通大学学报（自然科学版）, 2023, 42(8): 132-138.
[5]	丁天明，潘宁，杜柏松，艾万政. 基于改进灰色马尔可夫的港口货物吞吐量预测研究[J]. 重庆交通大学学报（自然科学版）, 2023, 42(9): 130-136.
[6]	常祎妹1，林忆婷1，丁天明2. 基于灰色马尔科夫模型的温州港货物吞吐量预测[J]. 重庆交通大学学报（自然科学版）, 2024, 43(11): 60-67.

基于CPSO算法改进GM-Markov模型的港口货物吞吐量预测

Port Cargo Throughput Forecasting Based on CPSO Algorithm Improved GM-Markov Model

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 6

编辑推荐

Metrics