基于BP神经网络的小汽车纵坡路段运行碳排放率预测

Prediction of Operational Carbon Emission Rate of Car based on BP Neural Networkin Longitudinal Slope Highway Section

随着高速公路里程规划和建设发展,公路行业节能减排目标的提出,如何实现对公路碳排放的控制和预测是研究的热点和难点问题之一。在广州增派公路设立试验路段进行现场行车试验,通过车载OBD设备获取油耗数据并换算成碳排放量。根据已有线形资料和路段划分,选取了弯坡路段平、纵线形9个相关参数作为影响因素,以碳排放率作为预测目标建立公路小客车弯坡路段行驶碳排放率多因素BP神经网络预测模型,进行了模型训练和预测性能检验。结果表明,所建立的BP神经网络模型在碳排放率预测方面具有较高的准确率,且其预测性能显著优于传统的多元线性回归模型。

With the continuous increase of highway mileage planning and construction,energy conservation and emission reduction targets of highway industry have been proposed.How to control and forecast highway carbon emissions has become a hot and difficult issue in current research.By selecting Zengpai Highway in Guangdong Province to set up test sections for the field driving test,and fuel consumption data has been obtained by on-board OBD equipment and converted into carbon emissions.According to the existing linear data and road section division,nine related parameters of horizontal and vertical curve of the curve section have been selected as the influencing factors.The multi-factor BP neural network prediction model has beenestablished with carbon emission rate as the prediction target.The model training and prediction performance test have been carried out.The results have shown that the BP neural network model has high accuracy in the prediction of carbon emission rate,and its prediction performance is significantly better than the traditional multiple linear regression model.