欢迎访问《汽车安全与节能学报》，
汽车安全与节能学报

汽车安全与节能学报 ›› 2025, Vol. 16 ›› Issue (1): 117-126.DOI: 10.3969/j.issn.1674-8484.2025.01.012

• 汽车节能与环保 • 上一篇    下一篇

车辆急动度对神经网络油耗预测性能影响研究

张立成(), 押境田, 彭琨, 杨冉   

  1. 长安大学 信息工程学院，西安 710064，中国
  • 收稿日期:2024-09-03 修回日期:2024-10-11 出版日期:2025-02-28 发布日期:2025-03-04
  • 作者简介:张立成（1987—），男（汉），江苏，高级工程师。E-mail：lichengzhang@chd.edu.cn。
  • 基金资助:
    “车-路信息感知与智能交通系统”111高校创新引智基地项目(B14043);中央高校基本科研业务费专项资金资助项目(CHD300102240503);中央高校基本科研业务费专项资金资助项目(CHD300102244501)

Impact of Jerk on neural network based fuel consumption predicting models

ZHANG Licheng(), YA Jingtian, PENG Kun, YANG Ran   

  1. School of Information Engineering, Chang’an University, Xi’an 710064, China
  • Received:2024-09-03 Revised:2024-10-11 Online:2025-02-28 Published:2025-03-04

RichHTML

2

PDF

226

摘要：

为了研究精细驾驶行为对基于单个和混合神经网络的油耗模型预测性能的影响，选择车辆急动度（Jerk）作为神经网络训练输入的重要变量。采用长短期记忆网络（LSTM）、循环神经网络（RNN）、非线性自回归带外部输入模型（NARX）、广义回归神经网络（GRNN）、卷积神经网络（CNN）、门控循环单元（GRU）、多层感知机（MLP） 以及卷积神经-长短期记忆网络（CNN-LSTM）混合神经网络共 8 种典型神经网络模型，选取（速度，加速度）、（速度，加速度和Jerk）、（发动机转速）共 3 种输入参数组合，以及校园低速、城市中速和高速公路高速共3种速度工况，累计进行了 69 组实验。结果表明：相较其余 6 种单个神经网络模型，LSTM 模型在各输入组合和各速度工况下的预测性能最好；CNN-LSTM混合模型的预测性能略优于 LSTM 模型。引入车辆急动度（Jerk）后，各神经网络油耗预测模型在各速度工况的预测性能都得到显著提高，其中，单个模型中，RMSE 最高下降了43.2%（CNN网络，高速路况），RE 最高下降了68.2%（LSTM网络，城市路况），R2 最高提升了 41.8%（NARX网络，城市路况）；混合模型中，RMSE 和 RE 分别最高下降了 34.9% 和 61.0%（城市路况）。

关键词: 生态驾驶, 油耗预测, 神经网络模型, 驾驶行为, 车辆急动度（Jerk）

Abstract:

In order to investigate the impact of refined driving behavior on the predictive performance of fuel consumption models based on individual and hybrid neural networks, the vehicular jerk was introduced as one crucial input variable in the training process. A total of eight typical neural network models were employed, including Long Short-Term Memory (LSTM), Recurrent Neural Network (RNN), Nonlinear Autoregressive model with Exogenous inputs (NARX), Generalized Regression Neural Network (GRNN), Convolutional Neural Network (CNN), Gated Recurrent Unit (GRU), Multilayer Perceptron (MLP), and Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) hybrid networks. Three input parameter combinations, namely (speed, acceleration), (speed, acceleration, and Jerk) and (engine speed), were selected, and 3 working scenarios, namely the low-speed campus scenario, the medium-speed urban scenario, and the high-speed expressway scenario, were selected, resulting in a total of 69 experiments setups. The results show that, among the seven individual neural network models, the LSTM model demonstrates the best predictive performance across all input combinations and driving conditions. The CNN-LSTM hybrid model exhibited slightly superior predictive performance compared to the LSTM model. Including vehicle jerk significantly improved the predictive performance of all neural network-based fuel consumption models across different speed conditions. Among the individual models, the Root Mean Square Error (RMSE) decreased by up to 43.2% (CNN model, highway condition), the Relative Error (RE) decreased by up to 68.2% (LSTM model, urban condition), and the coefficient of determination (R2) improved by up to 41.8% (NARX model, urban condition). In the hybrid models, RMSE and RE decreased by up to 34.9% and 61.0%, respectively (urban condition).

Key words: eco-driving, fuel consumption prediction, neural network model, driving behavior, vehicular Jerk

中图分类号: