基于ARIMA-LSTM与RBF-NOA的车速工况预测  

Vehicle Speed Condition Prediction Based on ARIMA-LSTM with RBF-NOA

作　　者：陆政元 杨昌波 李俊 鲍家定 郑伟光[1,2,3] Lu Zhengyuan;Yang Changbo;Li Jun;Bao Jiading;Zheng Weiguang(School of Electromechanical Engineering,Guilin University of Electronic Science and Technology,Guilin 541004,China;Commercial Vehicle Technology Center,Dongfeng Liuzhou Automobile Co.,Ltd.,Liuzhou 545005,China;School of Mechanical and Automotive Engineering,Guangxi University of Science and Technology,Liuzhou 545616,China)

机构地区：[1]桂林电子科技大学机电工程学院,广西桂林541004 [2]东风柳州汽车股份有限公司商用车技术中心,广西柳州545005 [3]广西科技大学机械与汽车工程学院,广西柳州545616

出　　处：《专用汽车》2025年第4期45-48,共4页Special Purpose Vehicle

基　　金：广西创新驱动发展专项基金项目(桂科AA23062040,桂科AA23062073);柳东科技计划项目(20210117);广西交通科技推广项目(GXJT-ZDSYS-2023-03-03)。

摘　　要：汽车能量的分析和管理离不开具体的行驶工况条件,因此车辆工况的构建和预测显得尤为重要。针对商用车,对典型车速工况进行数据处理后,首先利用ARIMA模型捕捉车速时间序列的趋势和周期性变化,判断时序数据是否具有季节性,然后,分别利用ARIMA-LSTM和NOA-RBF完成时间序列预测。结果表明,两种组合算法的应用提高了预测的精度和鲁棒性,降低了预测时间,为后续的能量管理与分析、控制以及智能驾驶奠定了坚实基础。The analysis and management of automobile energy cannot be separated from the specific driving conditions,so the construction and prediction of vehicle conditions are particularly important.In this paper,for commercial vehicles,after data processing of typical vehicle speed working conditions,the ARIMA model is firstly used to capture the trend and periodic change of vehicle speed time series,and to judge whether the time series data is seasonal or not,and according to the judgment,the time series prediction is completed by using ARIMA-LSTM and NOA-RBF respectively.The application of the two combined algorithms improves the accuracy and robustness of the prediction,reduces the prediction time,and lays a solid foundation for the subsequent energy management and analysis,control,and intelligent driving.

关 键 词：商用车 双模式工况预测 平稳性 

分 类 号：U461[机械工程—车辆工程]