The invention provides a genetic programming hyper-heuristic traffic flow distribution method based on knowledge migration. The method comprises the following steps: acquiring road network information; abstracting a road network into a weighted directed graph; constructing a traffic simulation model according to the road network information and determining an optimization target; evaluating each routing strategy in the population by using a traffic simulation model; performing knowledge extraction according to the source domain; when a problem scene changes, iterative optimization is carried out on a target domain based on knowledge learned from a source domain, and an optimal individual is used as a final routing strategy to be applied to traffic flow distribution. According to the method, the problem of effectiveness of traffic flow distribution by a routing strategy when a road network structure is changed is effectively solved, and meanwhile, the retraining efficiency in a new scene is improved; according to the method, transfer learning is applied to traffic flow optimization, general identification learned in a source field is transferred to a target field, the retraining efficiency under a new road network structure can be well improved, the method is suitable for traffic flow distribution strategies of different types of road network structures, and the generalization ability of a routing strategy is improved.

    本发明提供一种基于知识迁移的遗传规划超启发式交通流分配方法,包括获取路网信息;将路网抽象为带权有向图;根据路网信息构建交通仿真模型并确定优化目标;利用交通仿真模型对种群中每一个路由策略进行评估;根据源域进行知识提取;当问题场景发生变化时,基于从源域学到的知识对目标域进行迭代优化,将最优个体作为最终的路由策略应用于交通流分配。本发明有效解决面对路网结构发生变化时路由策略对交通流分配的有效性问题,同时提高了新场景下再训练的效率;将迁移学习运用于交通流优化中,将在源领域学到的通用识迁移到目标领域,能较好地提高在新路网结构下再训练的效率,适用于不同类型路网结构的交通流分配策略,提高路由策略的泛化能力。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Genetic programming hyper-heuristic traffic flow distribution method based on knowledge migration


    Weitere Titelangaben:

    一种基于知识迁移的遗传规划超启发式交通流分配方法


    Beteiligte:
    HU XIAOMIN (Autor:in) / LI JIAXUAN (Autor:in) / LI MIN (Autor:in)

    Erscheinungsdatum :

    19.11.2024


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    Hyper-heuristic traffic flow optimization method and system based on genetic programming

    HU XIAOMIN / DUAN YUHUI / LI MIN | Europäisches Patentamt | 2023

    Freier Zugriff

    HYPER-HEURISTIC STRUCTURAL OPTIMISATION OF CONCEPTUAL AIRCRAFT DESIGNS

    Allen, J. / Coates, G. / Trevelyan, J. et al. | British Library Conference Proceedings | 2012


    Hyper-Heuristic Structural Optimisation of Conceptual Aircraft Designs

    Allen, Jonathan / Coates, Graham / Trevelyan, Jon | AIAA | 2012



    Emergency Railway Transportation Planning Using a Hyper-Heuristic Approach

    Zheng, Yu-Jun / Zhang, Min-Xia / Ling, Hai-Feng et al. | IEEE | 2015