基于神经网络PID控制的单缸插销伸缩系统研究

发布时间：2018-02-28 17:01

本文关键词： 单缸插销液压系统神经网络PID控制伸缩臂起重机　出处：《大连理工大学》2011年硕士论文　论文类型：学位论文

【摘要】：单缸插销式伸缩臂技术是伸缩臂起重机特有的关键技术之一。随着伸缩臂起重机不断的向大型化、复杂化方向的发展,系统的非线性、不确定性、时变性、外界干扰和交叉耦合干扰对伸缩系统的位置精度的影响越来越明显,传统的PID控制方式难以达到控制要求。而神经网络PID控制器能够依据辨识网络提供的信息,实现控制参数的在线自适应调整,可以解决具有非线性、时变性的电液比例伸缩系统的控制问题,实现系统的智能控制。本文以大连理工大学与合肥神马重工合作开发的SMQ1000A伸缩臂起重机为为工程实例,对电液比例伸缩系统的智能控制方式进行了深入的研究,主要工作如下： (1)研究多种伸缩机构的结构特点和伸缩原理,设计了单缸插销式伸缩臂的电液控制系统,并对SMQ1000A (?)申缩臂起重机伸缩系统进行了匹配计算和元件选型； (2)基于AMESim对系统中关键液压元件,如电液比例阀、伸缩油缸等进行建模,在此基础上建立了单缸插销式伸缩臂伸缩机构液压系统仿真模型； (3)在Matlab/Simulink中应用单神经元模型建立了神经网络PID控制器,实现了电液比例伸缩系统的智能控制。在Matlab/Simulink中应用BP神经网络建立了神经网络辨识器,利用神经网络逼近任意非线性函数的能力,来模拟实际系统的输入—输出关系,两者有机结合构成本论文所设计整体控制器的核心部分； (4)联合仿真的研究及结果分析。将AMESim软件中建立的液压模型调用到Matlab/Simulink环境下,和本文设计的神经网络控制系统建立起电液闭环控制,实现了液压-控制的电液比例伸缩系统联合仿真。仿真结果表明,本文设计的神经网络PID控制系统,可以解决具有非线性、时变性的电液比例伸缩系统控制问题。仿真分析了影响油位置缩精度的主要因素,为设计提出了一些具有实际参考价值的建议。本文建立了单缸插销式伸缩臂伸缩机构液压系统的仿真模型,并与本文设计的神经网络PID控制系统进行了联合仿真,实现了系统的智能控制,其研究成果对改进工程机械电液比例伸缩系统的控制方式具有重要理论和实际意义。
[Abstract]:The single - cylinder bolt telescopic boom technology is one of the key technologies of the telescopic boom crane . As the telescopic boom crane continuously develops into large - scale and complicated direction , the nonlinear , uncertain , time - denatured , external disturbance and cross - coupling interference of the system are more and more obvious to the position accuracy of the telescopic system . The SMQ1000A telescopic boom crane developed by Dalian University of Science and Technology and Hefei Shenma Heavy Industry Co . , Ltd . is an example of engineering , and the intelligent control mode of the electro - hydraulic proportional telescopic system is studied deeply . The main work is as follows : ( 1 ) The structure characteristics and the principle of telescoping mechanism are studied , the electro - hydraulic control system of single - cylinder bolt telescopic boom is designed , and SMQ1000A ( ? ) is designed . Matching calculation and component selection are carried out on the telescopic system of the extension arm crane ; ( 2 ) modeling the key hydraulic components in the system , such as electro - hydraulic proportional valve , telescopic oil cylinder and so on based on the Sim Sim , and then establishing a simulation model of the hydraulic system of the single - cylinder bolt type telescopic arm telescopic mechanism ; ( 3 ) Using the single neuron model in Matlab / Simulink to establish the neural network PID controller , the intelligent control of the electro - hydraulic proportional telescopic system is realized . In Matlab / Simulink , the BP neural network is applied to establish the neural network identifier , and the input - output relationship of the real system is simulated by using the neural network to approximate any nonlinear function . The simulation results show that the neural network PID control system designed by this paper can solve the problems of nonlinear and time - denatured electro - hydraulic proportional telescopic system . Simulation results show that the neural network PID control system designed by this paper can solve the problems of nonlinear and time - denatured electro - hydraulic proportional telescopic system control . The simulation model of the hydraulic system of the telescopic mechanism of single - cylinder bolt - type telescopic arm is established , and combined with the neural network PID control system designed in this paper , the intelligent control of the system is realized , and the research results have important theoretical and practical significance for improving the control mode of the electro - hydraulic proportional telescopic system of engineering machinery .

【学位授予单位】：大连理工大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH212;TP183

【引证文献】