基于遗传算法优化的汽车电泳涂装输送用混联机构的分数阶PI~λD~u控制

发布时间：2018-05-11 20:15

本文选题：混联机构 + 耦合作用　；参考：《江苏大学》2017年硕士论文

【摘要】：现有汽车电泳涂装输送系统,如RoDip输送机和多功能穿梭机等皆采用悬臂梁结构,存在着承受重、大型负载能力较差,柔性化水平不高的问题。针对上述问题,本课题组在国家自然科学基金项目(51375210)的资助下,基于混联机构研制了一种新型汽车电泳涂装输送用混联机构。该混联机构兼具串联机构和并联机构的特点,其结构简单、承载能力强、柔性化水平高,应用领域更广。新型输送用混联机构以并联机构为主体,由多条运动支路组成,在机构运行过程中,由于具有高度非线性、强耦合、时变参数等特性,并存在未知扰动等,其运动控制较为复杂。经典运动控制策略如PID控制,其结构简单、易于操作和实现,在实际工程和生产设备的运动控制中被广泛使用,但对于模型不确定、非线性和多变量的被控对象,PID控制往往难以实现高性能控制。分数阶PI Dul控制器通过在整数阶PID控制器中引入积分阶次l和微分阶次μ,增大了控制器参数的整定范围,从而可得到更好的控制效果。此外,分数阶PI~λD~u控制器对其参数和被控对象系统参数的变化皆具有不敏感性,只要相关参数有限定的变化范围,分数阶PI~λD~u控制器能较好地实现控制。然而对于各支路间存在耦合的汽车电泳涂装输送用混联机构来说,分数阶PI~λD~u控制策略只能直接用于各支路的独立控制,因此需进一步考虑机构各支路之间的耦合问题。又由于分数阶PI~λD~u控制器引入了额外参数l和μ,因此还需考虑控制器参数优化的问题。为此,为解决输送用混联机构各支路间存在耦合的问题,本文提出了一种结合前馈补偿的分数阶PI~λD~u控制策略,并采用遗传算法实现对分数阶PI Dul控制器参数的优化。本文首先对汽车电泳涂装输送设备、混联机构及其运动控制策略进行了概述,并阐述了分数阶PI Dul特点及发展应用和遗传算法的发展概况。接着,针对新型输送用混联机构,建立运动学逆解、正解模型和雅可比矩阵,并通过MATLAB进行了运动学仿真,仿真结果表明了运动学分析的正确性。其次,采用Lagrange法建立新型输送用混联机构的动力学模型,在此基础上,进一步对输送用混联机构进行耦合性分析,并采用MATLAB对动力学模型以及输送用混联机构的耦合性进行了仿真,仿真结果表明了所建立模型的正确性以及输送用混联机构各支路间存在耦合作用的情况。然后,为解决输送用混联机构各支路间存在耦合的问题,设计了一种结合前馈补偿的分数阶PI~λD~u控制策略,即分别采用分数阶uDPIl控制对输送用混联机构各支路进行独立控制,同时通过建立动力学模型,进一步分析得出各支路间存在的耦合作用,采用前馈补偿的方式消除耦合作用的影响,并利用遗传算法优化控制器的参数。将所设计的控制器与未采用前馈补偿方式消除各支路间耦合作用影响的分数阶PI Dul控制器和PID控制器进行了MATLAB仿真比较,仿真结果表明结合前馈补偿的分数阶PI~λD~u控制器对外部干扰有着较强的鲁棒性,并且比未采用前馈补偿的分数阶PI~λD~u控制器和PID控制器具有更好的跟踪性能。最后,采用“PC+UMAC”的分布式结构,构建了新型输送用混联机构控制系统的实验平台,并完成新型输送用混联机构运动控制实验,从而进一步验证了本文所设计的结合前馈补偿分数阶PI~λD~u控制的可行性和有效性。
[Abstract]:The existing automobile electrophoretic coating transportation systems, such as RoDip conveyor and multi-functional shuttle machine, are used in cantilever beam structure. There is a problem of heavy load, poor large load capacity and low flexibility. Under the support of the National Natural Science Foundation Project (51375210), a kind of hybrid mechanism has been developed. The new type of hybrid mechanism has the characteristics of tandem mechanism and parallel mechanism. Its structure is simple, the bearing capacity is strong, the flexibility is high and the application field is more wide. The new type of mixing mechanism is mainly composed of parallel mechanism and is composed of multiple moving branches. In the operation process of the mechanism, because of the height of the mechanism Nonlinear, strong coupling, time-varying parameters and other characteristics, and there are unknown disturbances, its motion control is more complex. The classical motion control strategy, such as PID control, is simple in structure, easy to operate and realize, and is widely used in the motion control of actual engineering and production equipment, but for the model uncertain, nonlinear and multivariable controlled objects, PID Control is often difficult to achieve high performance control. The fractional order PI Dul controller can increase the tuning range of the controller parameters by introducing the integral order L and the differential order mu in the integer order PID controller, thus the better control effect can be obtained. In addition, the fractional order PI~ D~u controller changes the parameters of the controller and the parameters of the controlled object system. It is insensitive that the fractional order PI~ lambda D~u controller can achieve better control as long as the related parameters have a limited range of change. However, for the coupling mechanism of coupling between the branches of each branch, the fractional order PI~ lambda D~u control strategy can only be used directly for the independent control of various branches, so further consideration should be given to the mechanism. In addition, the fractional PI~ lambda D~u controller introduces the additional parameters L and mu, so the optimization of the controller parameters must be considered. In order to solve the problem of coupling between the branches of the mixing mechanism, a fractional order PI~ lambda D~u control strategy combined with feedforward compensation is proposed, and the relic is adopted in this paper. The transmission algorithm is used to optimize the parameters of the fractional order PI Dul controller. First, the paper summarizes the automobile electrophoretic coating transport equipment, the mixing mechanism and its motion control strategy, and expounds the characteristics of the fractional order PI Dul, the development and application and the development of the genetic algorithm. The kinematics simulation of the positive solution model and Jacobi matrix is carried out by MATLAB. The simulation results show the correctness of the kinematic analysis. Secondly, the dynamic model of a new type of mixing mechanism is established by Lagrange method. On this basis, the coupling analysis of the mixing mechanism is further analyzed, and the dynamic model is used by MATLAB. The simulation results show the correctness of the model and the coupling between the various branches of the mixing mechanism and the coupling between the various branches of the mixing mechanism. Then, in order to solve the problem of coupling between the branches of the mixing mechanism, a fractional order PI~ lambda D~u combined with the feedforward compensation is set up. The control strategy is to use the fractional uDPIl control to control each branch of the mixing mechanism separately. At the same time, through the establishment of the dynamic model, the coupling effect between the branches is further analyzed. The effect of the coupling is eliminated by the feedforward compensation method, and the parameters of the controller are optimized by the genetic algorithm. The controller is compared with the fractional order PI Dul controller and the PID controller which can eliminate the influence of the coupling interaction between the branches without the feedforward compensation. The simulation results show that the fractional PI~ D~u controller combined with the feedforward compensation has a strong robustness to the external interference, and the fractional order P without the feedforward compensation is compared. The I~ lambda D~u controller and the PID controller have better tracking performance. Finally, using the distributed structure of "PC+UMAC", the experimental platform of a new type of hybrid mechanism control system is constructed, and the motion control experiment of a new type of mixing mechanism is completed, which further validates the design of this paper combined with the feed forward compensation fractional PI~ [PI~] D~ The feasibility and effectiveness of u control.

【学位授予单位】：江苏大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U468.2;TP273

【参考文献】