基于数字孪生的重型车辆行驶状态预测研究

Prediction and Research on Vehicle Driving State Based on Digital Twin Models

为了详细研究重型车辆的行驶状态,利用两种不同的方法建立了车辆的数字孪生模型对其进行预测研究,并对孪生模型的适用范围进行了分析。首先利用相关装置和仪器对车辆的行驶参数和状态进行了测量,然后分别利用高斯过程和深度卷积神经网络建立了车辆行驶的数字孪生模型,两个模型的输入均为车辆的传动系统参数、动力系统参数及天气状况,输出参数为车辆的行驶速度和转矩值。分析结果显示,基于高斯过程的数字孪生模型对于车辆行驶参数的预测精度较高,基于深度卷积神经网络的孪生模型在短时间内的收敛精度较好。文中所建立的车辆数字孪生模型为后续的车辆行驶状态的优化及孪生交互技术的实现奠定了基础。

In order to study the driving state of heavy vehicles in detail,two different methods are used to establish a digital twin model of the vehicle for prediction research,and the applicable scope of the twin model is analyzed.Firstly,the driving parameters and state of the vehicle are measured using related devices and instruments,and then a digital twin model of the vehicle driving is established using Gaussian process and deep convolutional neural network.The input of the two models is the transmission system parameters of the vehicle.Power system parameters and weather conditions,the output parameters are the vehicle's driving speed and torque value.The analysis results show that the digital twin model based on Gaussian process has higher prediction accuracy for vehicle driving parameters,and the twin model based on deep convolutional neural network has better convergence accuracy in a short time.The vehicle digital twin model established in this paper lays the foundation for the subsequent optimization of vehicle driving state and the realization of twin interaction technology.