汽车磁流变液-弹耦合碰撞缓冲系统不确定时滞动力学行为及仿人自适应控制

项目名称： 汽车磁流变液-弹耦合碰撞缓冲系统不确定时滞动力学行为及仿人自适应控制

项目编号： No.60804018

项目类型： 青年科学基金项目

立项/批准年度： 2009

项目学科： 无线电电子学、电信技术

项目作者： 董小闵

作者单位： 重庆大学

项目金额： 20万元

中文摘要： 汽车碰撞不仅会损坏车体，严重时还会危及司乘人员安全，发展智能缓冲器是提高碰撞安全性并降低损失的有效途径。不少学者研究了调节弹性或阻尼的缓冲系统，但受限于执行器的缓冲机理和响应能力，效果并不理想。为此，项目以汽车磁流变液-弹耦合碰撞缓冲系统为对象，研究了不确定时滞对系统动力学行为的影响规律及仿人自适应控制方法。研究了基于磁流变弹性体变刚度原理以及高速冲击下基于磁流变液的缓冲器设计方法，对关键结构参数进行了优化设计；建立了汽车与固定壁碰撞的非线性动力学模型，运用非线性动力学理论研究了不确定时滞引起的复杂Hopf分叉力学行为，明确了控制参数的选择范围；以系统非线性动力学行为研究结果为基础，针对受控系统时滞不确定特点，研究了基于神经网络在线辨识的仿人自适应控制方法；通过仿真和跌落试验考查了磁流变缓冲器和控制算法的有效性。研究结果表明，采用磁流变缓冲器的缓冲隔振系统，在计算机仿真和跌落模拟汽车碰撞的试验中均能有效降低冲击过程的最大载荷；由于时滞的存在，汽车碰撞非线性动力学系统呈现复杂的分叉行为，通过合理选择控制参数可避免。项目的研究为开发先进的汽车碰撞智能缓冲系统进行了有益的理论探索和技术实践。

中文关键词： 汽车碰撞；磁流变；仿人智能控制

英文摘要： The vehicle crash has a very big impact on the damage of vehicle body and passengers' safety. An intelligent absorbing system is an effective method to promote the satefy and reduce the damage. Many researchers have studied the absorbing systems with variable stiffness and damping. However, the performance is not promising due to the working principle and the response time of absorbing systems. Therefore, a vehicle collision absorbing system based on magnetorheological fluids (MRF) and magnetorheological elastomer (MRE) is studied in this project. The law of uncertainty time delay on systems dynamics and the human simulated intelligent control (HSIC) are researched. The principle of variable stiffness of MRE is analzed while the design method of absorber based on MRF is proposed. The important design parameters are optimized. The nonlinear dynamic model of vehicle collision between the vehicle and the fixed barrier is developed. On the basis of theory of nonlinear dynamics, the Hopf bifurcation caused by the uncertain time delay is derived. The range of control parameters is determined. According to the characteristics of uncertain time delay, the HSIC based on neural network is proposed. The effectivity of the MR absorber and the control algorithm is validated by the numerical simulation and the drop tower test. The results show that the biggest impact load can significantly be reduced by the MR absorbing system during the impact course. Due to the time delay, the nonlinear dynamic system of vehicle collision has complex bifurcation dynamic behavior. The dangerous dynamic behavior can be avoided by determing rightly the control parameters. The research results of this project will provide the theoretical and technical foundation for developing advanced vehicle collision absorbing system.

英文关键词： vehicle collision;magnetorheological;human simulated intelligent control