分布式驱动HEV自适应无迹卡尔曼车速估计

发布时间：2017-12-31 09:41

本文关键词：分布式驱动HEV自适应无迹卡尔曼车速估计　出处：《中国科学:技术科学》2016年05期 　论文类型：期刊论文

【摘要】：针对四驱混合动力汽车,为改善其电子稳定性系统(electronic stability program,ESP)车速估计的精度和鲁棒性,考虑后轮毂电机转矩精确可测可控,以及前轮驱动转矩可推算的特点,结合ESP系统既有的传感器输出信号,提出了分布式自适应无迹卡尔曼车速估计算法.首先,建立了四驱混合动力汽车动力系统及动力学模型,其包括驱动系统模型、7自由度车辆动力学模型和Burckhardt轮胎模型.其次,考虑模型的时变及强非线性特性,采用无迹卡尔曼滤波(unscented Kalman filter,UKF)算法设计了主/子滤波器.一方面通过对量测噪声的自适应,改善了UKF算法量对量测信号干扰的鲁棒性;另一方面将主/子滤波器结果进行融合,并用融合后结果重置了各滤波器,提高了车速估计精度.最后,搭建了四驱混合动力汽车Carsim-Simulink联合离线仿真平台和硬件在环仿真试验平台,分别开展了双纽线、低速双移线两种驱动工况的离线仿真和硬件在环仿真试验.试验结果表明,所提出的分布式自适应无迹卡尔曼车速估计算法不仅估计精度较高,而且也有着较强的自适应性和鲁棒性能.
[Abstract]:For 4WD hybrid electric vehicle, in order to improve the electronic stability system (electronic stability program, ESP) the accuracy and robustness of speed estimation, after considering the wheel motor torque can be measured accurately controlled, and the characteristics of the front wheel drive torque can be calculated, combined with the existing ESP system sensor output signal, we propose a distributed adaptive unscented Calman speed estimation algorithm. Firstly, a four-wheel drive hybrid powertrain system and dynamics model, which includes a drive system model with 7 degrees of freedom vehicle model and Burckhardt tire model. Secondly, considering variable and strong nonlinear characteristics of the model, using the unscented filter Calman (unscented Kalman filter UKF) algorithm to design the main / sub filter on the one hand. On the adaptive measurement noise, improve the robustness of UKF measurement on the amount of signal interference; on the other hand will be the main / sub filter For the fusion of each filter and reset by fusing the results, improve the speed estimation accuracy. Finally, build a 4WD HEV Carsim-Simulink combined with off-line simulation platform and hardware in the loop simulation test platform, carried out lemniscate, low-speed double shift line two driving the off-line simulation and hardware in the loop simulation test. The test results show that the proposed distributed adaptive unscented Calman speed estimation algorithm not only high estimation accuracy, but also has strong adaptability and robustness.

【作者单位】：同济大学新能源汽车工程中心;同济大学汽车学院;
【基金】：国家自然科学基金联合基金(批准号:U1564208)资助项目
【分类号】：U469.7;U463.6;U467
【正文快照】： 引用格式:赵治国,朱强,周良杰,等.分布式驱动HEV自适应无迹卡尔曼车速估计.中国科学:技术科学,2016,46:481 492Zhao Z G,Zhu Q,Zhou L J,et al.Vehicle speed estimation in driving case based on distributed self-adaptive unscented kalman filter for 4WD hybrid*E-mail

【参考文献】