节点文献
电梯用直线感应电机的优化设计及其控制系统研究
Research on Optimization Design and Control System of Linear Induction Motor Used in Elevator
【作者】 贾宏新;
【作者基本信息】 浙江大学 , 电气工程, 2002, 博士
【摘要】 本文所研究的直线电机电梯是具有良好发展前景的新型电梯。直线电机电梯结构简单,无需机房,甚至不需要绳索和对重,因此占地面积少,另外还有高速、高层、可靠等优点,适用于超高层建筑和七层以下的建筑,是电梯的重要发展方向。在世界上,其研究和应用刚刚起步,国内还没有研究。 本文从驱动电梯用圆筒型直线感应电动机的电磁设计开始,再到直线电机的优化设计;接下来设计了包括操作和驱动在内的直线电机电梯数字控制系统,并研究了直线电机电梯的控制方式和驱动策略。 本文主要做了如下几个方面的工作: 第一章回顾了直线电机的发展历程,总结了直线电机的理论分析方法和设计特点,简要介绍了直线电机电梯的优点和不同结构类型,确立了本文的研究主题:电梯用直线感应电动机的优化设计及其控制系统研究。 第二章从圆筒型直线感应电动机电磁设计入手,讨论了结构和参数的选择对性能的影响,重点论述了直线电机设计准则——品质因数的设计问题,得出不同性能要求对应不同的品质因数值,设计中要综合考虑。 第三章为得到更好的力能指标,对直线电机进行优化设计。首次把一种新型全局优化算法——区域消去法引入到电机优化设计中。针对电机优化问题的具体特点,提出将混合罚函数法和修正的鲍威尔法相结合用于其中的局部寻优。应用该算法对直线电机进行优化设计,较大提高了直线电机的力能指标。 第四章设计了直线电机电梯全数字控制系统,该系统能够减少系统的开关触点个数,提高系统的可靠性,并能实现复杂的控制策略,提高电梯运行的效率和舒适度。采用最新的DSP芯片TMS320F240和MCS51单片机,结合分布式控制理论,将整个电梯的控制功能分为若干模块,分别由不同的控制器来实现,提高了系统效率,减少了井道布线,易于扩展和检修。 第五章针对直线电机特有的边缘效应,分析了动态纵向边缘效应的影响,它产生端部涡流损耗,减小气隙磁场。这种效应可以表示成直线电机初级铁心有效长度、次级电阻、次级电感和速度的函数。把这种效应考虑到按转子磁链定向的dq0同步旋转等效电路中去,建立了直线感应电动机的矢量控制数学模型。仿真结果表明,对动态纵向边缘效应进行补偿后,系统的动态特性得到提高。本章还把智能控制手段应用于直线电机调速系统,提出了增量式单神经元PI速度调节算法,这样的单神经元控制器可视为由神经元实现的自适应的非线性PID控制器,能够实现比传统PID控制器更好的性能。 第六章总结了全文的主要研究成果,并展望了下一步的研究工作。 本文所做工作的创新性,主要表现在以下方面: 研究和设计对象——直线电机驱动的电梯是国内第一台直线电机电梯模型。在设计过程中,分析了不同的品质因数对性能的影响,将全新的全局优化算法——区域消去法应用于直线感应电动机的优化设计,将混合罚函数法和修正的鲍威 浙江大学博士学位论文 尔法相结合用于其中的局部寻优,取得了较好的效果;设计了基于分布式控制理 论的全数字直线电机电梯控制系统,采用DSP芯片TMS320F240和MCS51设计 了主控制器和副控制器,主控制器对电梯的各种运行状态及其驱动直线电机的变 频调速进行控制,各楼层门厅和轿厢分别由各自的MCS51组成的副控制器进行控 制,主控制器与各副控制器之间采用RS—485方式进行串行通讯;针对直线电机 的特点,把动态纵向边缘响应的影响等效成电机参数的函数,考虑到矢量控制的 模型中去,建立了直线感应电动机较准确的控制模型;本文还设计了增量式单神 经元PI速度调节器,具有较高的自适应能力和实用性。’
【Abstract】 The linear-motor-driven elevator (LMDE), introduced in this paper, is one new kind of promising elevator. LMDE can be used in both skyscrapers and buildings less than seven floors, which has the following advantages: its structure is simple, and it need no penthouse machine room, even no suspension cables and counterweight, thus space-saving. It works with high speed and security, and can be applied to high floors. The research and application of LMDE is rising in the world, however up to now, there are no domestic reports.The electromagnetic design of tubular linear induction motor (TLIM), driving elevator, is the first part of this paper. Then optimum algorithm of TLIM is studied. In the next part, full digital control system of LMDE, including command system and drive system, is discussed. The research of control method and drive algorithm is the last part of this paper.The main content of this paper is as follows:In chapter 1, the development of linear motor is reviewed, and the theoretical analyses and design characteristics of the linear motor are generalized. According to the brief introduction of the advantages and various structures, the research subject is set to optimize TLIM for the drive of elevator and its controller design.In chapter 2, based on the electromagnetic design of TLIM, the influence of TLIM’s construction and parameters on its properties is discussed. Giving emphasis to goodness factor design, one can conclude that the value of goodness factor must be chosen carefully to get different performance.In chapter 3, in order to improve the product of the power factor and efficiency of TLIM, a new algorithm for global optimization, the Domain Elimination Algorithm (DBA), is used in the optimal design. The Mult-SUMT method is combined with improved POWELL method to conduct local optimization procedure. As optimal results, the main index has a considerable improvement.In chapter 4, the full digital control system of elevator, using DSP chip TMS320F240 and microcontroller MCS51, is designed, which can reduce switch counts, improve reliability, achieve complex control algorithm, and obtain high efficiency and great comfort of riding. The distributed control architecture for elevator is presented. The control system is divided into several function modules which are controlled by different controller modules, so the efficiency is improved, the hoistway lines are reduced, and it is easy for people to expand and repair the control system.In chapter 5, the dynamic vertical end effect of linear motor is analyzed, which causes eddy current loss and air gap flux decrease. This effect can be represented by thefunction of primary length, speed, secondary inductance and resistance. An equivalent circuit model in the rotor flux oriented frame is developed for the vector control of TLIM by adding the end effect function. The simulation results show the improvements achieved by the proposed scheme. An intelligent control strategy, incremental PI speed controller based on single neuron, is presented, which can be regarded as the adaptive nonlinear controller and can achieve more satisfactory performance than traditional PID controllers.In chapter 6, all achievements of the dissertation are summarized and the further research work is foreseen.The creativity of this dissertation is:The research object, the linear-motor-driving elevator, is the first domestic model. During the design process, the influence of goodness factor on motor’s performance is discussed, a new global optimal algorithm-domain elimination algorithm is applied in TLIM’s design, Mult-SUMT method and improved POWELL method are used as local optimization. The optimal results are perfect. Based on distributed control architecture, the full digital controller of LMDE is designed using DSP chip-TMS320F240 and microcontroller MCS51. The main controller controls the elevator’s statuses and drives the TLIM using VVVF method; auxiliary controllers, including floor controllers and car controller, are designed by MCS51. The
【Key words】 linear induction motor; optimal design; domain elimination algorithm; linear-motor-driven elevator; vector control; single neuron;