基于RBF网络PID控制的单缸插销伸缩系统研究
发布时间：2018-10-21 12:35
【摘要】：单缸插销伸缩臂作为全地面起重机的关键部件,在很大程度上决定了整机的起重性能,也是全地面起重机控制的关键技术之一。随着起重机发展的智能化、大型化,被控对象的严重非线性、时变参量较多、时变大、控制精度要求高等特点越来越明显,常规的数字PID控制技术已经不能很好的处理这类问题,RBF神经网络整定的PID控制通过对被控对象动态特性的辨识,能很好地为其匹配对应的数学模型,并根据数学模型的规律实时在线整定PID参数,可以满意地解决上述问题,具有优良的控制精度。 本文以大连理工大学与太原重工股份有限公司合作开发的TZM500全地面起重机为课题来源,对单缸插销伸缩系统的控制方法进行了对比研究,主要进行的研究工作如下： (1)归纳了国内外伸缩臂结构的主要类型,并对各自的工作原理进行分析,结合全地面起重机的特点,对其单缸插销伸缩臂的电液控制系统方案进行设计,对其中涉及到的外购件进行了选型匹配。 (2)在软件AMESim中建立了伸缩臂液压系统模型,并加入AMESim/Simulink接口,为联合仿真做准备。 (3)分析常规增量式PID控制、BP网络整定的PID控制、RBF网络整定的PID控制的仿真效果的各项指标,在Matlab/Simulink中建立了RBF网络整定的PID控制器模型。 (4)将AMESim模型以S函数的形式调入Matlab/Simulink模型,进行联合仿真,验证控制器的控制效果,对结果进行分析。 本文设计了RBF网络整定的PID控制器,并将其应用到伸缩臂液压模型的控制中,结果表明RBF网络整定的PID具有很好的控制效果,提升了单缸插销式伸缩臂的控制精度。
[Abstract]:As a key part of all-ground crane, single-cylinder pin telescopic boom determines the lifting performance of the whole crane to a great extent, and is also one of the key technologies in the control of all-ground crane. With the development of intelligent and large-scale crane, it is more and more obvious that the controlled object is seriously nonlinear, with many time-varying parameters, large time-varying, high control precision and so on. The conventional digital PID control technology has not been able to deal with this kind of problem well. The PID control based on RBF neural network can match the corresponding mathematical model well through the identification of the dynamic characteristics of the controlled object. According to the rule of mathematical model, the parameters of PID can be adjusted in real time and on-line, which can solve the above problems satisfactorily and have good control precision. Based on the TZM500 all-ground crane developed by Dalian University of Technology and Taiyuan heavy Industry Co., Ltd., the control methods of single-cylinder pin expansion system are compared and studied in this paper. The main research works are as follows: (1) the main types of telescopic boom structures at home and abroad are summarized, and their working principles are analyzed, combined with the characteristics of all-ground cranes. The electro-hydraulic control system scheme of the telescopic arm of the single cylinder pin is designed, and the type selection of the purchased parts is carried out. (2) the hydraulic system model of the telescopic arm is established in the software AMESim, and the AMESim/Simulink interface is added. Preparation for joint simulation. (3) Analysis of the conventional incremental PID control, BP network tuning PID control, RBF network tuning PID control simulation results of the indicators, The PID controller model of RBF network tuning is established in Matlab/Simulink. (4) the AMESim model is transferred into the Matlab/Simulink model in the form of S function, and the joint simulation is carried out to verify the control effect of the controller, and the results are analyzed. In this paper, the PID controller of RBF network tuning is designed and applied to the control of the hydraulic model of the telescopic arm. The results show that the PID tuned by the RBF network has a good control effect and improves the control accuracy of the single-cylinder pin telescopic arm.
【学位授予单位】：大连理工大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH131.6
本文编号：2285105
[Abstract]:As a key part of all-ground crane, single-cylinder pin telescopic boom determines the lifting performance of the whole crane to a great extent, and is also one of the key technologies in the control of all-ground crane. With the development of intelligent and large-scale crane, it is more and more obvious that the controlled object is seriously nonlinear, with many time-varying parameters, large time-varying, high control precision and so on. The conventional digital PID control technology has not been able to deal with this kind of problem well. The PID control based on RBF neural network can match the corresponding mathematical model well through the identification of the dynamic characteristics of the controlled object. According to the rule of mathematical model, the parameters of PID can be adjusted in real time and on-line, which can solve the above problems satisfactorily and have good control precision. Based on the TZM500 all-ground crane developed by Dalian University of Technology and Taiyuan heavy Industry Co., Ltd., the control methods of single-cylinder pin expansion system are compared and studied in this paper. The main research works are as follows: (1) the main types of telescopic boom structures at home and abroad are summarized, and their working principles are analyzed, combined with the characteristics of all-ground cranes. The electro-hydraulic control system scheme of the telescopic arm of the single cylinder pin is designed, and the type selection of the purchased parts is carried out. (2) the hydraulic system model of the telescopic arm is established in the software AMESim, and the AMESim/Simulink interface is added. Preparation for joint simulation. (3) Analysis of the conventional incremental PID control, BP network tuning PID control, RBF network tuning PID control simulation results of the indicators, The PID controller model of RBF network tuning is established in Matlab/Simulink. (4) the AMESim model is transferred into the Matlab/Simulink model in the form of S function, and the joint simulation is carried out to verify the control effect of the controller, and the results are analyzed. In this paper, the PID controller of RBF network tuning is designed and applied to the control of the hydraulic model of the telescopic arm. The results show that the PID tuned by the RBF network has a good control effect and improves the control accuracy of the single-cylinder pin telescopic arm.
【学位授予单位】：大连理工大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH131.6
【参考文献】
相关期刊论文 前3条
1 林培群;徐建闽;傅惠;梁俊斌;;多分支BP网络模型及其在车型分类中的应用[J];微计算机信息;2005年25期
2 姬晓飞,孟令柏,申东日,陈义俊;基于RBF神经网络多步预测的自适应PID控制[J];甘肃科学学报;2003年02期
3 宫赤坤,华泽钊;RBF网络在线辨识的神经元控制[J];自动化仪表;2002年11期
相关硕士学位论文 前1条
1 刘益民;基于改进BP神经网络的PID控制方法的研究[D];中国科学院研究生院（西安光学精密机械研究所）;2007年
，本文编号：2285105
本文链接：https://www.wllwen.com/kejilunwen/jixiegongcheng/2285105.html
