磁流变减振器极限高低温特性及变温正逆模型

磁流变减振器阻尼力和电流的精确控制是实现半主动悬架的算法、达到整车系统控制目标的必要条件，但由于磁流变液的温度敏感性使得磁流变减振器阻尼力强烈依赖温度变化，带来模型失配和控制效果弱化的问题。基于此进行磁流变减振器在不同电流和速度下的高低温(-40℃~80℃)示功试验研究,揭示磁流变减振器在低温下丧失阻尼特性而表现出刚度特性，在高温下黏滞阻尼退化的特性规律。为了描述磁流变减振器随温度变化的复杂非线性特性，提出一种新的磁流变减振器变温参数化双曲滞回模型，该模型引入温度作为自变量，对磁流变减振器黏滞阻尼、刚度及滞回特性进行准确描述。为了面向实际减振器控制，在此双曲滞回模型的基础上，进一步线性化求逆得到磁流变减振器温度修正的逆模型。该逆模型输入预期阻尼力和减振器压缩速度作为自变量，可以直接给出满足减振器力值约束的控制电流。研究结果表明：相较于未进行温度补偿的逆模型，该逆模型能够平均提升12.79%的电流控制精度以及12.53%的控制力跟踪精度；进行温度修正的模型能够在仿真中还原更真实的磁流变减振器力学特性，逆模型能够给出更精确的控制电流，为充分发挥磁流变减振器的能力、实现车辆的半主动悬架精确控制提供了理论和方法指导。

Magnetorheological Damper Extreme High and Low Temperature Characteristics and Variable Temperature Forward and Reverse Models

Accurate control of magnetorheological damper (MRD) damping force and current is necessary to realize the algorithms for semi-active suspensions and to achieve the control objectives of the whole vehicle system. However, the temperature sensitivity of magnetorheological fluids makes MRD damping force strongly dependent on temperature variations, bringing about problems of model mismatch and degradation of control effects. In this paper, an experimental study is carried out to show the performance of MRD at different currents and velocities, at high and low temperatures from -40 ℃ to 80 ℃. It is revealed that the MRD loses damping characteristics at low temperatures and exhibits stiffness characteristics, while the viscous damping degrades at high temperatures. In order to describe the complex non-linear characteristics of the MRD with temperature, a parametric hyperbolic hysteresis model of the MRD is proposed, which introduces temperature as an independent variable to accurately describe the viscous damping, stiffness, and hysteresis characteristics of the MRD with an RMSE below 0.15 at different temperatures. The temperature-revised inverse model for the MRD is further linearized to obtain. The inverse model needs the demanding damping force and the damper compression velocity as independent variables to present a direct control current satisfying the force constraints of the damper. The results show that the inverse model validated by experimental data could improve the current control accuracy by an average of 12.79% and the control force tracking accuracy by 12.53% compared to the inverse model without temperature revision. The temperature-revised model is able to reproduce more realistic magnetorheological damper mechanics in the simulation, and the inverse model is able to perform more accurate control currents, providing theoretical and methodological guidance to fully exploit the capability of MRD and realize accurate control of semi-active suspensions in vehicles.