LQR最优控制器在精馏塔上的应用

本文将ＬＱＲ最优控制的理论研究成果应用到精馏塔上，设计了精馏塔的线性二次型最优控制器，并将控制结果与常规ＰＩＤ控制器控制结果进行了比较。仿真证明，在动态变化过程中，ＬＱＲ控制器确实能使目标函数得到优化。     

精馏模糊解耦控制系统探究

本文介绍了精馏塔的精馏工艺,分析了精馏塔塔顶和塔底温度控制回路的耦合关系,提出了模糊解耦控制应用的必要性。针对精馏塔塔顶和塔底温度的控制特点,设计了一种双输入双输出模糊解耦控制系统,具体分析了系统中该模糊解耦控制器的结构以及模糊控制各环节的具体实现方法,应用Simulink对控制系统进行仿真研究,得到了较好的控制效果。     

精馏过程的神经网络控制系统的设计

本文以一个三元精馏塔为研究对象，详细介绍了精馏过程的神经网络控制系统ＣＮＮＣＳ的设计过程，本ＮＮＣＳ主要包括ＮＮ邓估器和ＮＮ控制器，前者被训练后用来精确表达精馏塔的输入到输出的非线性映射，后者被训练后用来对该ＮＮ预估器实现解耦控制，所采用的神经网络模型为ＢＰ网络，是在一个以ＰＣ机为宿主机带９片ＴＲＡＮＳＰＵＴＥＲ的并行处理系统上实现的。     

控制结构约束下状态反馈LQR系统的设计

线性二次型调节器问题（即ＬＱＲ问题）最优状态反馈律的实现必须要求全部状态变量可测，而实际工程应用中常常难以保证。针对此问题，文中引入控制结构约束，使反馈只充分应用可测状态变量，在此基础上本文对终端时间限ｔｆ＝∞的ＬＱＲ系统进行了分析，推导论证了相应的状态调节器的必要条件，并给出了一个实现算法。     

浅谈JX-300X控制系统在精馏塔控制中的应用

精馏塔是很多工业蒸馏过程中最重要的设备,精馏的好坏将直接影响到产品的质量和工厂的效益,因此对它的控制自从一出现就受到业内的关注。本文针对在乙醇生产中怎样利用SUPCONJX-300X系统对精馏塔进行控制提出了一个应用方案。     

基于双模糊控制器的水箱液位控制

针对工业锅炉自动控制系统中的水箱系统液位控制问题,提出基于模糊控制器的基础上,设计实现一种双模糊控制器。根据输出信号的误差大小分别利用两个模糊控制器进行控制,控制结果传递给调节器,以实现水箱水位不变。从仿真结果来看,双模糊控制器有效地减小了系统稳态误差,响应时间、超调量、稳定时间等性能优于传统的PID控制。     

基于发动机平均值模型的控制仿真

针对发动机平均值模型进行控制仿真以验证所设计的控制策略对发动机控制的有效性,首先在MATLAB／Simulink中建立发动机的平均值模型,其次分别建立PID控制器和模糊PID控制器对发动机平均值模型进行控制仿真,最后对两种控制器的仿真结果进行比较分析。通过最后的仿真结果可以看出两种控制器都可以稳定控制发动机模型,但模糊PID控制器具有更好的控制效果。发动机是一个非线性、时滞的动力学系统,而模糊控制不需要了解被控对象的特性,故它能精确控制发动机模型。     

空间机器人全局逼近收敛控制

针对自由漂浮空间机器人系统,设计了一种全局逼近收敛控制器,解决了存在初始状态误差及外界干扰条件下的系统鲁棒性问题。首先,根据空间机器人系统的状态空间方程设计控制器,该控制器将控制输出特性与控制增益隔离开,对系统模型的不确定性及外界干扰具有强鲁棒性。另外,该控制方法避免了类似反演控制方法带来的复杂度高的问题。其次,从理论上证明了该控制器能够实现机器人系统对期望路径的跟踪。最后,针对单臂六自由度空间机器人系统,分别使用该控制器与鲁棒补偿控制器进行了仿真分析。仿真结果证实了该控制器的强鲁棒性。     

基于PIC芯片嵌人式电机控制器的研究

提出一种基于PIC芯片的嵌入式电机控制器。该控制器最大的特点在于位置、速度、电流三种控制方式可随时变更,主要用于多轴服务机器人的伺服控制。着重介绍该控制器的软件构造、操作指令和通信模式。在控制器软件的开发上,采用模块化的设计思想,并且设计了使用方便的指令模式和通信错误处理机制。位置控制、速度控制和电流控制的试验结果指出其控制精度能够满足多轴服务机器人的需要。     

两轮移动倒立摆的开关切换模糊极点配置控制器设计

为了更好地控制两轮移动式倒立摆,文章以两轮移动式倒立摆动力学数学模型为基础,设计了一种开关切换模糊极点配置控制器,利用该控制器对系统进行控制。通过一个选择型开关把模糊控制和状态反馈的极点配置控制有机的统一起来构成了开关切换模糊极点配置控制器。在Simulink中,对设计的开关切换模糊极点配置控制器进行了仿真,实现了两轮移动式倒立摆的平衡、匀速前进、匀速转弯。仿真结果说明了开关切换模糊极点配置控制器对两轮移动式倒立摆起到一定的抗干扰和理想的控制作用。     

Piezoelectric vibration control for all-clamped panel using DOB-based optimal control

Considering the spillover and harmonic effect in real active vibration control, a novel composite controller based on disturbance observer (DOB) for the all-clamped panel is presented. The single-mode of the piezoelectric panel can be regarded as a second-order system. The unmodeled error of the current controlled mode, harmonic effects, uncontrolled mode effects, etc., are regarded as the lumped disturbances which can be estimated by the DOB, and the estimated value is used for the feed-forward compensation design. Then, an optimal linear quadratic regulator (LQR) strategy is employed for the feedback design. In order to solve the difficulty of determining the weight matrices of LQR, a chaos optimization method based on logistic map is proposed. So the weight matrices can be tuned automatically. Combining with a new transient performance function, the optimal weight matrices can be obtained. The composite controller can effectively suppress the lumped disturbances of the all-clamped panel. Experiment comparisons with conventional LQR are given to verify the effectiveness of the proposed method.     

线性二次型调节器算法研究

LQR 即线性二次型调节器,其对象是现代控制理论中以状态空间形式给出的线性系统,而目标函数为对象状态和控制输入的二次型函数。LQR的最优解可以写成统一的解析表达式, 且可导致一个简单的状态反馈控制律。线性二次型控制理论已成为反馈设计系统的一种重要工具, 其特点是：为多变量反馈系统的设计提供了一种有效地分析方法;它可以适应于时变系统;它可以处理扰动信号和测量噪声问题;它可以处理有限和无限的时间区间。在处理可精确数学描述的对象时取得了很大的成功。对研究无人机的自动飞行控制技术具有十分重要的现实意义。     

A sensorless optimal control system for an automotive electric power assist steering system

This paper considers the analysis and design of a double-pinion-type electric power assist steering (EPAS) control system. A simplified model of the augmented steering assembly-electric motor system is developed using Lagrangian dynamics, and an optimal controller structure for the model is proposed. Three main advances to the state of the art are presented in this paper. First, a state-space design model is used rather than an input-output model. A state-space formulation for a system model that incorporates motor electrical dynamics is obtained with the assist motor angular position as the output. Second, linear quadratic regulator (LQR) and Kalman filter techniques are employed to arrive at an optimal controller for the EPAS system. The selection of weighting coefficients for the LQR cost function is discussed. Finally, the authors present a control strategy that eliminates the steering column torque sensor, a critical component in existing EPAS controller designs. The proposed control strategy presents an opportunity to improve EPAS system performance and also reduce system cost and complexity.     

Digital Modeling and PID Controller Design for MIMO Analog Systems with Multiple Delays in States, Inputs and Outputs

This paper presents a discrete-time state-space methodology for the digital modeling and design of an optimal digital proportional-integral-derivative (PID) plus state-feedback controller for multiple-input, multiple-output (MIMO) continuous-time systems with multiple time delays in states, inputs and outputs. To implement the digital design, first the Chebyshev quadrature formula together with a linear interpolation method is employed to obtain an extended discrete-time state-space model from the continuous-time multiple time-delay system. Then, a partially predetermined digital PID controller and the extended discrete-time state-space model are formulated as an augmented discrete-time state-space system utilizing state-feedforward and state-feedback linear-quadratic regulator (LQR) design. As a result, the parameters of the optimal PID controller and its associated state-feedback controller can be determined by tuning the weighting matrices in the LQR performance criteria. Further, an optimal discrete-time observer is jointly constructed for the multivariable system with multiple delays in states, inputs and outputs. The proposed design methodology can be applied to general MIMO continuous-time multiple time-delay systems for performance improvement and disturbance rejection. An illustrative example is given to demonstrate the effectiveness of the developed method. This work was supported in part by U.S. Army Research Office under Grant W911NF-06-1-0507, the National Science Foundation under Grant NSF0717860, and Research Contract 1440234.     

A salient object detection framework using linear quadratic regulator controller

In this paper, a novel salient object detection framework based on Linear Quadratic Regulator (LQR) controller is proposed. The major goal of this research is to take advantage of optimal control theory for improving the performance of detecting salient objects in images. In this regard, for the sake of detection of salient and non-salient regions, two LQR-based control systems are employed. In the proposed framework, for the initialization of the control systems, background and foreground estimations have been done with two different strategies. Doing so, we would ultimately have more effective distinction between those regions. After the initialization step, the control systems refine both estimations in parallel until reaching a steady state for each of them. Within the mentioned process, by using optimal control concept, specifically LQR controller (for the first time in the field), control signals which are in charge of determining saliency values, would be constantly optimized. At the end, the raw saliency map will be generated by combination of background and foreground optimized initial maps. Finally, the integrated saliency map will be refined by using angular embedding method. The experimental evaluations on three benchmark datasets shows that the proposed framework performs well and introduces comparable results with some deep learning based methods.     

Ultra-lightweight anthropomorphic dual-arm rolling robot for dexterous manipulation tasks on linear infrastructures: A self-stabilizing system

This paper proposes the application of a very low weight (3.2 kg) anthropomorphic dual-arm system capable of rolling along linear infrastructures such as power lines to perform dexterous and bimanual manipulation tasks like the installation of clip-type bird flight diverters or conduct contact-based inspection operations on pipelines to detect corrosion or leaks. The kinematic configuration of the arms, with three joints at the shoulder and one at the elbow, allows the natural replication of the human movements to conduct these tasks, exploiting also the kinematic redundancy of the shoulder to maintain the equilibrium while perching on the line. The dynamic model of the system is derived to design a self-stabilizing controller that maintains the base of the arms at an equilibrium point. The state-dependent Riccati equation (SDRE) controller is chosen for this purpose since the system is under-actuated and the contribution of the control gain (with nonlinear optimal structure) on all states is critical. The SDRE is a nonlinear optimal controller that extends the margins of stability in comparison with linear ones. Simulation results show that the SDRE performs the regulation to the equilibrium point successfully and evidence better performance with respect to a linear quadratic regulator (LQR). The system is validated in an outdoor testbed consisting of a power line mockup, presenting experimental results to evaluate the SDRE and LQR controllers, demonstrating also the autonomous installation of clip-type bird flight diverters and the aerial deployment using a multirotor platform.     

An improved LQR-based controller for switching DC-DC converters

A general approach for controlling pulse-width-modulated (PWM) -type switching DC-DC converters digitally using state-feedback techniques and linear optimal control theory is reported. The methodology for redesigning the state estimator is investigated, and a method derived from the general linear-quadratic-regulator (LQR) problem, is proposed. The method is found to offer better transient responses and robustness to uncertainties in plant parameters when compared with the typical eigenvalue-assignment method. Special attention is directed to plant models with possible migrations of the open-loop zeroes across the stability boundary during operation. Results of applying these techniques to a published Cuk converter are reported to illustrate different points of interest     

LQR在单级倒立摆系统中的应用

单级倒立摆控制是一个即复杂而又对准确性、快速性要求很高的非线性不稳定系统控制问题。单级倒立摆数学模型的建立对研究其稳定性具有指导作用。针对多变量、非线性、强耦合性的倒立摆系统,运用牛顿动力学方法建立其动力学方程,并进行线性化处理,得到状态空间模型。然后对该模型分别进行LQR控制,在MATLAB环境下进行仿真。实验结果表明,二次型最优控制具有良好的响应性能和算法简单等特点,在实际应用中具有重要意义。     

The control of switching DC-DC converters-a general LWR problem

The control of switching DC-DC converters is reviewed. It is regarded as a general linear quadratic regulator (LQR) problem, and an innovative optimal and robust digital controller is proposed. The control strategy adopted can achieve good regulation, rejection of modest disturbances, and the ability to cater to switching converters with RHP zeros. This controller design is a general approach that is applicable to all PWM-type DC-DC converters with their circuit topologies known or unknown. Modern CAD techniques are used to reach the final control law. Application to a published Cuk converter is used as an example, and the performance is evaluated     

Linear Quadratic Control of a Loaded Agonist?Antagonist Muscle Pair

A discrete-time control law which gives the optimal applied force for a given position trajectory of a linear second-order external load is computed using linear quadratic regulator (LQR) theory. This control law is used with an existing first-order fixed interpulse interval (IPI) muscle force controller to regulate position in a simulated electrically stimulated loaded agonist-antagonist muscle system. Both force and position feedback are required to implement the control strategy. A nonlinear second-order model of a predominantly slow twitch muscle is used in the simulation. The results suggest that good position control of a loaded antagonist-agonist muscle system is possible if a linear model of the external load is appropriate.