伺服系统的分析与辨识

发布时间：2018-05-28 01:36

本文选题：系统辨识 + 两步法　；参考：《西安电子科技大学》2014年硕士论文

【摘要】：本文主要内容是对机载雷达转台伺服系统和复合轴伺服系统的粗转台模型进行研究与辨识。系统辨识是控制领域研究的重点之一,因为精确的动力学模型是进行机电系统设计与控制的基础,在控制领域具有相当重要的地位。利用系统辨识的相关理论知识,通过对被控对象结构和参数的辨识,建立被控对象精确的数学模型,这是研究控制系统的关键步骤,然后基于此模型进行控制算法的研究,从而使其满足控制系统的指标要求。本论文主要进行了两个伺服系统的辨识工作,其中包括基于实验室环境下的机载雷达转台伺服系统的辨识,还有基于实际科研项目中的复合轴伺服系统中粗转台的辨识工作。其中机载雷达转台伺服系统部分的工作主要包括四个方面：辨识实验方案设计；辨识模型结构的确定；辨识模型参数的确定；辨识模型的验证。其中辨识实验方案设计的工作主要包括了根据转台伺服系统的原理,参与设计了系统的软硬件实现方案,采集输入输出数据,为被控对象模型的辨识工作做准备。输入信号为STM32产生的M序列,输出信号为采集到的编码器的速度信号。模型结构的辨识主要采用AIC准则推导出系统阶数。模型参数的辨识主要采取两步法来进行辨识,首先利用相关法得到系统的非参数模型——脉冲响应,而后又利用最小二乘法推算出系统的参数模型——传递函数。模型结构和模型参数确定之后,我们对系统模型的有效性进行验证,主要方法为输入信号给定,将模型输出曲线与实际采集输出曲线进行比较,误差小于一定的指标要求,实验证得辨识模型真实可靠。由于机载雷达转台伺服系统需要在特殊环境稳定运行,如高低温环境、振荡环境等,需要在被控对象参数发生一定变化时系统仍然能够可靠运行。故后期基于辨识模型又进行了PID参数自整定的研究,主要为基于三环路的模糊智能控制方法的研究,使系统具有一定的鲁棒性。最后基于上述工作,对系统进行了闭环调试,结果表明伺服系统各项功能正常,与上位机可以可靠地通信。复合轴转台伺服系统的的辨识工作是根据工程上给定的一组输入输出数据,对复合轴伺服系统的粗转台模型进行辨识。这组数据为一组椭圆航迹测试数据,经分析其主要成分为正弦信号,由于正弦信号无法在全频段充分激励系统,故我们对输入输出信号进行一阶差分和二阶差分运算,使其能够激励系统的各个模态。由于工程上存在噪声,我们应用最小二乘拟合法对输入输出信号进行滤波处理。根据处理后的这些输入输出数据,基于相关最小二乘两步法辨识出系统的各类模型。从这些模型中筛选出精度比较高的模型,并对模型的有效性进行了验证。复合轴伺服系统的精转台就是建立在这个粗转台之上的,因此对粗转台模型的辨识为后续精转台的设计提供了重要的依据。本文主要内容为对两个典型的伺服系统进行系统辨识工作,然后对辨识的模型均进行了有效性验证,结果表明模型误差均在要求的范围内,辨识模型真实可靠。
[Abstract]:The main content of this paper is to study and identify the rough turntable model of the airborne radar turntable servo system and the compound axis servo system. The system identification is one of the key points in the field of control research. Because the precise dynamic model is the basis for the design and control of the electromechanical system, it has a very important position in the control system. The identification of relevant theoretical knowledge, through the identification of the structure and parameters of the controlled object, establish the precise mathematical model of the controlled object. This is the key step in the study of the control system. Then based on this model, the control algorithm is studied to meet the requirements of the control system. This paper mainly carries out the identification of two servo systems. It includes identification of the airborne radar turntable servo system based on the laboratory environment, and the identification of the rough turntable in the composite axis servo system based on the actual scientific research project. The work of the airborne radar turntable servo system mainly includes four aspects: the identification experiment scheme design; the identification model structure The determination of the parameters of the identification model, the identification of the identification model, and the verification of the identification model. The work of the identification experiment scheme mainly includes the design of the software and hardware implementation scheme of the system according to the principle of the turntable servo system, the input and output data are collected, the identification work of the controlled object model is prepared. The input signal is the M produced by the STM32. The output signal is the speed signal of the encoder. The identification of the model structure mainly uses the AIC criterion to derive the order of the system. The identification of the model parameters is mainly adopted by the two step method. First, the non parametric model of the system is obtained by the correlation method, and then the system is calculated by the least square method. The parameter model - transfer function. After the model structure and model parameters are determined, we verify the validity of the system model. The main method is the input signal is given, the output curve of the model is compared with the actual acquisition and output curve, the error is less than the requirement of certain index. The model is proved to be true and reliable. Due to the airborne mine, the model is proved to be true and reliable. The servo system of the turntable needs to operate steadily in a special environment, such as high and low temperature environment, oscillating environment, and so on. The system is still able to operate reliably when the parameters of the controlled object have certain changes. Therefore, a study on the self-tuning of PID parameters based on the identification model is carried out in the later period, mainly for the study of the fuzzy intelligent control method based on the three loop. Finally, the system has a certain robustness. Finally, based on the above work, the closed-loop system is debugged. The results show that the functions of the servo system are normal and can communicate reliably with the upper computer. The identification work of the servo system of the composite axis turntable is based on a set of input and output data given in the project and the rough rotation of the servo system of the composite axis. This group of data is a group of elliptical track test data. After analyzing its main component is sine signal, because the sine signal can not fully stimulate the system in the whole frequency band, we carry out the first order difference and the two order difference operation to the input and output signal, so that it can stimulate the various modes of the system. Sound, we apply the least squares fitting method to filter the input and output signals. According to the processed data, we identify the various models of the system based on the correlation least square two step method. The model is selected from these models, and the validity of the model is verified. The fine turntable is built on this coarse turntable, so the identification of the rough turntable model provides an important basis for the subsequent design of the fine turntable. The main content of this paper is to identify the two typical servo systems, and then verify the validity of the identified models. The results show that the model errors are all in need. The identification model is true and reliable in the range of the calculation.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN959.73

【参考文献】