电子学报 ›› 2021, Vol. 49 ›› Issue (5): 879-886.DOI: 10.12263/DZXB.20200915

所属专题: 面向自动驾驶和智慧交通协同的通信与控制

• 面向自动驾驶和智慧交通协同的通信与控制 • 上一篇    下一篇

反向学习的灰狼算法优化及其在交通流预测中的应用

张兴辉1,3, 樊秀梅1, 阿喜达1,4, 樊书嘉2, 武文瑜1   

  1. 1. 西安理工大学自动化与信息工程学院, 陕西西安 710048;
    2. 清华大学计算机科学与技术系, 北京 100084;
    3. 安康学院电子与信息工程学院, 陕西安康 725000;
    4. 包头师范学院信息科学与技术学院, 内蒙古包头 014030
  • 收稿日期:2020-08-21 修回日期:2021-02-18 出版日期:2021-05-25
    • 通讯作者:
    • 樊秀梅
    • 作者简介:
    • 张兴辉 男,1982年4月生于辽宁省营口县.2009年获长春理工大学硕士学位,其后在安康学院从事教学工作,讲师.2018年至今在西安理工大学信息与控制系攻读博士学位.主要研究方向为车联网、智能交通控制. E-mail:1180311021@stu.xaut.edu.cn
    • 基金资助:
    • 国家自然科学基金 (No.61272509); 陕西省"百人计划"、陕西省重点研发计划一般项目 (No.2017ZDCXL-GY-05-01)

Grey Wolf Optimization Based on Opposition Learning and Its Application in Traffic Flow Forecasting

ZHANG Xing-hui1,3, FAN Xiu-mei1, SHAN Axida1,4, FAN Shu-jia2, WU Wen-yu1   

  1. 1. Department of Automation and Information Engineering, Xi'an University of Technology, Xi'an, Shaanxi 710048, China;
    2. Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China;
    3. College of Electronics and Information Engineering, Ankang University, Ankang, Shaanxi 725000, China;
    4. School of Information Science and Technology, Baotou Teachers' College, Baotou, Inner Mongolia 014030, China
  • Received:2020-08-21 Revised:2021-02-18 Online:2021-05-25 Published:2021-05-25
    • Corresponding author:
    • FAN Xiu-mei
    • Supported by:
    • National Natural Science Foundation of China (No.61272509); 100 Talents Project of Shaanxi Province,  General Program of Key Research and Development Project of Shanxi Province (No.2017ZDCXL-GY-05-01)

摘要: 本文针对灰狼优化(Grey Wolf Optimizer,GWO)算法平衡全局探索和局部搜索能力的不足,提出了一种基于反向改进的灰狼算法(Opposition Learning Grey Wolf Optimizer,OLGWO),来优化预测模型的超参数,以提高其用于交通流预测的精度与鲁棒性.本算法在迭代过程中采用了反向学习策略,并引入了等级相关概念,主要通过计算普通狼与目标狼的Spearman相关系数,并根据其值来选择性地更新狼种群.实验先对12个标准测试函数对比了四种算法OLGWO、TGWO(Transformed Grey Wolf Optimizer)、GWO、PSO(Particle Swarm Optimization),得到了寻优均值和标准差,验证了OLGWO算法具有突出的性能优势;然后采用美国加州公路交通流数据,在不同缺失率下比较了四种算法优化的反向传播(Back Propagation,BP)网络模型,结果显示,OLGWO-BP模型预测精度比其它三种模型最高分别有1.95%、3.98%和11.07%的提升,同时表现出更好的稳定性.

关键词: 智能交通, 交通流预测, 灰狼优化(GWO), BP神经网络, 反向学习(OL)

Abstract: Focusing on the weakness of the balance capability between global exploration and local search in the classical GWO algorithm, a novel grey wolf algorithm based on opposition learning (OLGWO), which can evolve the hyper-parameters of forecasting model, is proposed to improve the accuracy and enhance the robustness of traffic flow forecasting models. This algorithm is designed to take advantage of opposition learning strategy with the iterative process, and exploits the concept of rank correlation that can describe the Spearman correlation coefficients between the target wolf and the common wolves, and then selectively updates the each wolf of the whole population according to their values. Firstly, the performance comparison of four algorithms (OLGWO, TGWO, GWO, PSO), based on 12 benchmark functions, is conducted in terms of the two metrics, namely the optimization means and standard deviations. The results verify the outstanding performance of the proposed algorithm. Furthermore, based on the California highway traffic flow data, the four models optimized by the concerned algorithms are compared under different loss rates. The results show that the prediction accuracy of OLGWO-BP is higher than that of the others by 1.95%, 3.98% and 11.07%, respectively, and the stability is better.

Key words: intelligent transportation, traffic flow prediction, grey wolf optimizer (GWO), BP neural networks, opposition learning

中图分类号: