【作者】 靳宝全；

【导师】 熊诗波；

【作者基本信息】 太原理工大学 ， 机械电子工程， 2010， 博士

【摘要】 近年来,随着科技的发展,采用电液伺服控制的机械系统愈来愈复杂,并且存在非线性、参数不确定性、机械谐振、未建模动态特性、传感器动力学特性、外负载干扰等因素影响,对控制系统的精度、响应能力、稳定性及鲁棒性的要求则愈来愈高,系统的复杂性与苛刻的控制性能要求之间形成了尖锐的矛盾。滑模变结构控制是一种非线性控制器,当系统状态穿越状态空间的不同区域时,反馈控制器的结构按照一定的规律发生变化,使得控制系统对被控对象的内在参数变化和外部扰动具有一定的适应能力,保证系统的性能达到期望的品质。滑模变结构控制系统的鲁棒性要比一般常规的连续控制系统强。但是对于一个实际的滑模变结构控制系统,控制力受限、系统惯性、切换开关的时间与空间滞后、检测误差及离散化形成的准滑模等,都会造成抖振。抖振现象给变结构控制在实际系统中的应用带来了困难,因此对其控制信号抖振的消弱成为变结构研究的热点问题。在解决抖振问题的研究上,国内外研究者提出了许多方法,这些方法要么在消弱抖振的同时也降低了系统鲁棒性,要么系统过于复杂,无法应用到实际工程领域。因此,设计一种满足实时性、鲁棒性要求,并有效抑制抖振的先进滑模变结构控制策略具有重要的理论意义与工程应用价值。滑模变结构控制和模糊控制是各自独立发展起来的两类控制方法,二者都是对不确定系统进行控制的有效方法,它们的结合会进一步提高控制效果。本文针对滑模变结构控制存在的问题,设计模糊滑模变结构控制器,通过控制特性的互补来获得满意的控制性能。作者提出按照系统的实时性和鲁棒性的要求设计模糊滑模控制算法,对提出的方法进行理论分析、混合仿真以及系统的试验验证,从而为可靠的工程应用奠定理论和方法基础。电液控制系统影响因素复杂,不能用精确模型来描述其所有特性,所以,在控制系统的分析和设计中,往往采用简化模型。设计中被忽略的因素可能会引起控制系统品质的恶化、甚至导致不稳定。为提高建模精度,在对阀控缸电液位置伺服系统分析的基础上,考虑系统参数不确定性,建立了基于位置变量与偏差变量的系统状态空间模型,通过数字仿真初步确定控制器参数。为更接近实际系统,使用实际的液压-机械系统物理模型和数字控制器模型,实现阀控缸电液位置控制系统的混合仿真,从而建立一个更加接近实际控制状态的模型。分析了滑模变结构控制器设计的基本问题及其Lyapunov稳定性,阐述了模糊控制器的设计及其稳定性分析问题,并探讨了模糊理论与滑模变结构理论的几种结合方案。针对现有模糊滑模变结构控制策略的缺陷,提出对模糊滑模变结构控制进行更深入的研究,以满足实际工程应用要求。针对固定参数的趋近律滑模变结构算法无法根据系统参数的变化和干扰的变化进行实时调整的缺点,提出模糊控制器来调整趋近律参数的方法。基于模糊自适应指数趋近律函数切换滑模控制,对非线性、扰动和参数不确定性有较好的鲁棒性,并且克服了常规滑模抖振大和控制力频繁切换的缺点,且实时性强、控制精度高。将这种方法应用于某大型钢铁集团公司硅钢厂电液单辊CPC控制系统,并以其物理模型与模糊自适应指数趋近律函数切换滑模控制器模型,进行混合仿真。研究表明,系统在综合考虑非线性、扰动及各种参数不确定性的情况下,模糊自适应指数趋近律函数切换滑模控制能够稳定工作,有效地抑制了抖振。比例滑模策略保留了线性控制的某些优点,但抖振的存在不仅会降低控制精度,甚至会激发系统的未建模动态或引起机械谐振,这些不足严重制约了比例滑模变结构控制在大负载高精度电液位置系统中的应用。为避免抖振对系统精度与稳定性影响,提出了采用模糊模型,根据切换函数及其导数的状态自调整比例滑模切换增益,以柔化控制信号。作者研制的大型钢铁集团公司液压EPC大负载高精度位置伺服系统的混合仿真结果表明,通过模糊理论实现增益自调整,有效降低抖振,既实现了高精度控制,又保留了滑模策略抗参数摄动及抗扰动能力强的特点。最后以DSPACE平台设计了电液伺服综合试验系统,实现了基于模糊滑模控制策略的快速原型试验。构建了电液伺服综合试验系统的硬件,设计了基于结构不变性原理的电液位置系统加载策略。针对电液位置系统的非线性、参数不确定性及外力扰动,采用提出的基于模糊自适应趋近律函数切换滑模及模糊自调整增益比例滑模方法,进行了实时控制试验,通过与常规控制策略的比较,验证了所提策略的有效性。更多还原

【Abstract】 In recent years, with the development of technology, electro-hydraulic servo controlled mechanical systems increasingly become complex, and there are non-linear, parameter uncertainty, mechanical resonance, unmodelled dynamic charcteristics, sensor dynamic charcteristics, load disturbance factors, etc.And the control system accuracy, responsiveness, stability and robustness requirements are higher and higher, so that a sharp contradiction appears between the system’s complexity and demanding performance requirements of the control.Sliding mode variable structure control is a nonlinear controller.When the system state goes through the different areas of the state space, and the feedback controller structure changes according to certain rules,the control system has a certain ability to adapt the object system internal parameter variations and external disturbance,and ensure the system performance to achieve the desired quality. Sliding mode variable structure control system has better robustness than the conventional continuous control systems.But for a real sliding mode variable structure systems, it exists control limited, the system inertia switch in time and space lag, testing errors and the formation of discrete quasi-sliding mode and so on, which can cause buffeting. It made it difficult to the variable structure control in the application of the actual system, so the issue is to weaken the buffeting of variable structure for its control signal. During solving the problems, there are a number of ways, but the problem exists that either reduce the system robustness at the same time weaken the buffeting, or policy is too complex only for the computer simulation that can not be putted into practical engineering applications. Therefore, the design an advanced sliding mode variable structure strategy has very real significance that meets the real-time, robustness requirements, and effectively inhibits the buffeting.Sliding mode variable structure control and fuzzy control is developed independently of the two types of control methods, both are the effective methods for the uncertain control system, and their combination will further enhance the control effect. In this paper, according to the sliding mode variable structure control problems, fuzzy sliding mode controller was designed, by complementary controlling properties to obtain satisfactory control performance. In accordance with the real-time and robustness requirements, the fuzzy sliding mode algorithm was designed.By theoretical derivation, the hybrid simulation and test verification, it will lay the theoretical and methodological foundation for liable engineering applications.Influence factors of electro-hydraulic control system are very complex so accurate model is not described all features. In the analysis and design of system, simplified model is often used. Negligible factors may cause the quality deterioration and even instability of control system. On the basis of systematic analysis on electro-hydraulic servo valve control cylinder system, to improve modeling accuracy and consider the uncertainty of system parameters, system state-space model is established by using variables and deviation variables. For a more realistic system, using the actual hydraulic-mechanical system physical model and the digital controller model, the thesis achieves the hybrid simulation of electro-hydraulic valve-controlled cylinder position control system, so as establish a more realistic control state model.The thesis analyzes the basic problems and Lyapunov stability of sliding mode controller, describes the design and stability analysis of fuzzy controller, and investigates several programs of the fuzzy theory and sliding mode variable structure theory combination. For the deficiencies of fuzzy sliding mode variable structure control strategy, fuzzy sliding mode variable structure control for more in-depth study is proposed by the author in order to meet the practical engineering applications.The fuzzy controller is designed for adjusting the reaching law parameters, because the algorithm of reaching law sliding mode variable structure with fixed parameters can be not real-time adjusted with the system parameter and interfere changes. Switch sliding mode control of fuzzy adaptive reaching law index function has good robustness for non-linear, disturbance and parameter uncertainty, and overcomes the conventional sliding mode control big buffeting and frequent switch shortcomings. The system has stronger real-time and higher control accuracy. The control method is used in a large steel iron and steel group silicon electro-hydraulic single-roller CPC system, which the physical model and switch sliding mode controller model of fuzzy adaptive reaching law index function is for hybrid simulation. Studies show that the control system works stably and effectively suppresses chattering on the condition of the system comprehensive consideration nonlinearity, disturbance and parameters uncertainty.The proportion sliding mode strategy retains the linear control advantages, but the presence of buffeting will reduce the control precision, even stimulate the system unmodeled dynamics or cause mechanical resonance. These deficiencies seriously restricted the application of the proportion of sliding mode control in large-load high-precision electro-hydraulic position system. In order to avoid chattering influence on the system accuracy and stability, self-tuning sliding mode switching gain is designed to soften the control signal, according to the switch function and its derivative of the fuzzy model. As example of some iron and steel group EPC, which is a big load high precision hydraulic position servo system, is simulated hybrid. The results show that the system achieves the gains self-adjusting by fuzzy theory, effectively reduces the buffeting, implements high-precision control, and retains characteristics of the anti-parameter perturbations and anti-disturbance ability.Electro-hydraulic servo test system on Dspace is designed to achieve a fuzzy sliding mode control strategy based on rapid prototyping test. The electro-hydraulic servo test system hardware is designed, the system load policy on structural invariance principle was designed. For the electro-hydraulic position system of nonlinear, parameter uncertainty and external disturbances, by using proposed fuzzy adaptive reaching law and fuzzy self-tuning gain proportional sliding mode for real-time control, by comparison with the conventional control strategy to verify the validity of the proposed strategy.更多还原