双遗忘因子最小二乘法车辆质量和时变坡度估计  

Estimation of vehicle mass and time-varying slopefor the double forgetting factor recursiveleast square method

作　　者：叶明[1] 卢祥伟 张利杰 杨洲 周俊充 范毅 郑易 YE Ming;LU Xiangwei;ZHANG Lijie;YANG Zhou;ZHOU Junchong;FAN Yi;ZHENG Yi(Key Laboratory of Advanced Manufacturing Technology for Automobile Parts,Ministry of Education,Chongqing University of Technology,Chongqing 400054,China;Chongqing Jinkang Power New Energy Co.,Ltd.,Chongqing 401335,China)

机构地区：[1]重庆理工大学汽车零部件先进制造技术教育部重点实验室,重庆400054 [2]重庆金康动力新能源有限公司,重庆401335

出　　处：《重庆理工大学学报（自然科学）》2023年第6期1-9,共9页Journal of Chongqing University of Technology：Natural Science

基　　金：国家自然科学基金面上项目(52172400);重庆市技术创新与应用发展专项(cstc2021jscx-gksbX0043)。

摘　　要：针对双遗忘因子最小二乘法由于数据饱和导致车辆变质量估计失效,在坡道起步时导致坡度估计性能下降的问题,提出一种改进算法。建立了车辆行驶时的纵向动力学数学模型,并根据最小二乘递推估计理论得到双遗忘因子最小二乘递推估计模型;针对原始算法的变质量估计问题,通过引入车速作为车辆停车的判断参数,并在停车后重新初始化协方差矩阵来消除数据饱和的影响;采用AVL CRUISE与Matlab/Simulink联合仿真对识别算法进行了验证。仿真对比分析表明,有效地验证了改进算法的合理性和可靠性,提高了算法在车辆质量和坡度估计时的收敛速度和精度。This paper proposes an improved algorithm for the problem that the double forgetting factor recursive least square method leads to the failure of vehicle variable mass estimation due to data saturation and the decline of slope estimation performance when a vehicle starts on a ramp.Firstly,a longitudinal dynamics mathematical model of vehicle running is established and a double forgetting factor recursive least square estimation model is obtained based on the recursive least square estimation theory.Secondly,an improved algorithm is designed for the variable mass estimation of the original algorithm to eliminate the effect of data saturation by introducing vehicle speed as the judgment parameter for vehicle parking and re-initializing the covariance matrix after parking.Finally,the recognition algorithm is simulated and verified by using a joint simulation of AVL CRUISE and Matlab/Simulink.Through simulation and comparison analysis,the rationality and reliability of the improved algorithm are effectively verified,and the convergence speed and accuracy of the algorithm in vehicle mass and slope estimation are improved.

关 键 词：车辆质量估计 道路坡度估计 双遗忘因子最小二乘法 数据饱和 

分 类 号：U463.2[机械工程—车辆工程]