摘要
为了解决算法复杂度高导致仿真平台与实验模型脱节的问题,提出了一种基于灰色预测技术的智能交通虚实融合实验教学模型。该模型由可视化仿真平台和灰色校正模型组成。可视化仿真平台采用S7-200 SMART PLC作为控制器,并使用MCGS作为监控软件实现构建。通过构建OPC服务器,实时更新的车辆数据从PLC传输到MATLAB进行处理。进一步,提出了基于海洋捕食者算法优化的灰色伯努利模型,并将该模型应用于交通流量的滚动预测。预测结果被用作信号配时的依据,以实现交通信号灯的动态调节。实验结果表明,引入高效的灰色预测技术后,对交通流量序列的预测误差具有更小的数值,相对误差均控制在3%以下;优势分析发现预测模型对外部扰动的变化细节得到了更准确的反应,实验教学平台在可视化、实操性和智能化特征方面得到了显著改善。
In order to address the issue of the disconnection between simulation platforms and experimental models caused by high algorithm complexity,an intelligent traffic virtual-reality integration experimental model based on grey prediction technology is proposed.The model consists of a visualization simulation platform and a grey calibration model based on small data.The visualization simulation platform utilizes the S7-200 SMART PLC as the controller and employs MCGS as the monitoring software for construction.By establishing an OPC server,realtime updated vehicle data is transmitted from the PLC to MATLAB software.Furthermore,the grey Bernoulli model based on the Marine Predators Algorithm is proposed.This model is applied to rolling prediction of traffic flow,and the prediction results are used as the basis for signal timing to dynamically adjust traffic lights.The experimental results indicate that the introduction of efficient grey prediction technology leads to smaller prediction errors in traffic flow sequences,with relative errors controlled below 3%.Advantage analysis reveals that the prediction model provides more accurate responses to changes and disturbances in the external environment.The experimental teaching platform shows significant improvements in terms of visualization,practicality,and intelligence.
作者
李守军
Li Shoujun(School of Mechanical and Electrical Engineering,Suqian University,Suqian,Jiangsu 223800,China)
出处
《机电工程技术》
2024年第12期162-166,197,共6页
Mechanical & Electrical Engineering Technology
基金
宿迁市科技计划项目(Z2022097)
宿迁市智能制造重点实验室(M202108)
宿迁学院实验教学和教学实验室建设研究项目(2024SYJJ03)。
关键词
智能交通
灰色模型
MCGS
海洋捕食者算法
实验教学
intelligent transportation
grey model
MCGS
marine predator algorithm
experimental teaching
