基于磁流变阻尼结构的PID控制算法优化研究

发布时间：2018-06-06 08:54

本文选题：磁流变阻尼器 + PID控制　；参考：《南京理工大学》2017年硕士论文

【摘要】：磁流变阻尼器(Magnetorheological Fluid Dampers,简称MRFD)具有性能稳定、易于控制、响应速度快且出力大等优点,是一种在建筑结构的振动控制拥有广阔前景的新型减震装置。PID控制算法是振动控制中常用的控制算法,但PID控制缺乏自我调节能力,这也降低了其控制效果,因此研究基于磁流变阻尼结构的PID控制算法具有重要的意义。本文研究了磁流变阻尼器的设计过程,PID控制算法的优化以及阻尼器在建筑中的优化布置三个方面,主要的研究成果如下:(1)从阻尼器的材料、尺寸、磁路三个方面介绍本课题组自主研制的最大行程320mm、最大出力为20kN的剪切阀式磁流变阻尼器的设计过程。对阻尼器进行力学性能测试,获得磁流变阻尼器的阻尼力与加载幅值、频率及电流的关系,确定阻尼器的力学性能参数。(2)提出采用神经网络对PID控制进行优化改进的策略和采用模糊控制优化PID控制的策略,优化后的PID算法可根据地震加速度和结构自身的状态对其控制参数进行实时的调节,提升PID算法的振动控制效果。(3)将本文提出的神经网络优化PID控制、模糊控制优化PID控制与传统PID控制三种算法进行对比。三种控制算法用于某安装有磁流变阻尼器的九层框架结构的仿真分析中,仿真结果表明:在结构位移的控制效果上,神经网络优化PID控制和模糊控制优化PID控制均比传统PID控制算法有明显提升;在结构加速度的控制效果上,神经网络优化PID控制的提升更加明显。(4)为提高磁流变阻尼结构的经济性,改进了振型组合的性能指标的阻尼器优化布置方案,并与性能指标增量法确定的优化布置方案进行对比,仿真实验表明:对于结构的位移和加速度的控制上,改进振型组合的性能指标法优化效果均优于性能指标增量法,该性能指标切实有效。
[Abstract]:Magnetorheological Fluid dampers (MRFDs) have the advantages of stable performance, easy control, fast response speed and large force. Pid control algorithm is a new kind of vibration absorber with broad prospect in the vibration control of building structure. It is a commonly used control algorithm in vibration control. However, the PID control lacks the ability of self-regulation, which also reduces its control effect. Therefore, it is of great significance to study the PID control algorithm based on Mr damping structure. In this paper, the design process of magnetorheological damper is studied. The optimization of pid control algorithm and the optimal layout of damper in building are studied. The main research results are as follows: 1) the material and size of the damper, This paper introduces the design process of the shear valve magnetorheological damper with maximum stroke of 320mm and maximum force of 20kN developed by our group from three aspects of magnetic circuit. The mechanical properties of the damper are tested, and the relationship between the damping force and the loading amplitude, frequency and current of the Mr damper is obtained. Determining the mechanical performance parameters of damper. (2) the strategy of optimizing and improving PID control by neural network and the strategy of optimizing PID control by fuzzy control are put forward. The optimized PID algorithm can adjust the control parameters in real time according to the seismic acceleration and the state of the structure itself, and improve the vibration control effect of the PID algorithm. (3) the neural network proposed in this paper is used to optimize the PID control. Three algorithms of fuzzy control optimization PID control and traditional PID control are compared. Three control algorithms are applied to the simulation analysis of a nine-story frame structure with magneto-rheological damper. The simulation results show that the control effect of the displacement of the structure is obtained. Neural network optimal PID control and fuzzy control optimal PID control are significantly improved than the traditional PID control algorithm. In order to improve the economy of magnetorheological damping structure, the optimal layout scheme of damper for the performance index of modal combination is improved. Compared with the optimal layout scheme determined by the incremental performance index method, the simulation results show that the improved mode combination performance index method is superior to the performance index increment method in the control of displacement and acceleration of the structure. The performance index is effective.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TU311.3

【相似文献】