RBF-ARX model-based MPC strategies with application to a water tank system

A hybrid pseudo-linear RBF-ARX model that combines Gaussian radial basis function (RBF) networks and linear ARX model structure is utilized for representing the dynamic behavior of a class of smooth nonlinear and non-stationary systems. This model is locally linear at each working point and globally nonlinear within whole working range. Based on the structural characteristics of the RBF-ARX model, three receding horizon predictive control (RBF-ARX-MPC) strategies are designed: (1) the RBF-ARX-MPC algorithm based on single-point linearization (MPC-SPL); (2) the RBF-ARX-MPC algorithm based on multi-point linearization (MPC-MPL); and (3) the RBF-ARX-MPC algorithm based on globally nonlinear optimization (MPC-GNO). In the MPC-SPL, the future multi-step-ahead predictive output of the system is obtained based on the local linearization of the RBF-ARX model at only current working-point, while in the MPC-MPL the future long-term output prediction is obtained according to the future local characteristics from previous online optimization results of the RBF-ARX model based MPC. In the MPC-GNO, the globally nonlinear characteristics of the RBF-ARX model are fully used for online getting control variables of the MPC. Real-time control experiments for the three type MPCs are carried out on a water tank system, which are also compared with a classical PID control and a traditional linear ARX model-based MPC. The results verify that the modeling method and the model-based predictive control strategies are realizable and effective for the nonlinear and unstable system. Moreover, it is also shown that the MPC-GNO can obtain better control performance but need more computation time compared to the other MPCs, which makes it possible to be applied into some slowly varying processes.     

Distributed model predictive control for plant-wide hot-rolled strip laminar cooling process

Since hot-rolled strip laminar cooling (HSLC) process is a large-scale, nonlinear system, a distributed model predictive control (DMPC) framework is proposed for computational reason and enhancing the precision and flexibility of control system. The overall system is divided into several interconnected subsystems and each subsystem is controlled by local model predictive control (MPC). These local MPCs cooperate with its neighbours through the scheme of neighbourhood optimization for the improvement of global performance. The state space representation of each subsystem’s prediction model is designed by finite volume method firstly, and then is linearized around the current operating point at each step to overcome the computational obstacle of nonlinear model. Moreover, since the strip temperature is measurable only at a few positions in water cooling section due to the difficult ambient conditions, an Extended Kalman Filter (EKF) is used to estimate the transient temperature of strip. Both simulation and experiment results prove the efficiency of the proposed method.     

Reduction of the QP-MPC cascade structure to a single layer MPC

Some commercial MPC packages are implemented in two layers, the QP static layer and the MPC dynamic layer. In the absence of an upper Real Time Optimization layer, the static layer solves a simplified economic optimization problem, which defines optimum feasible targets for the dynamic layer. Since the LP/QP static layer and the MPC dynamic layer are usually executed within the same sampling period, it is not trivial to guarantee that the interaction between the two layers will not disrupt the stability of the whole structure. In this paper, it is proposed an approach to reduce the two-layer structure of some commercial MPCs to a single dynamic layer where the control cost function is extended to include the economic objective. In the proposed approach the convergence and stability of the closed-loop system can be obtained if the economic term of the cost function is properly weighted. A simulation example of a simple industrial system shows the efficiency of the proposed strategy.     

Feedback Linearization Based Distributed Model Predictive Control for Secondary Control of Islanded Microgrid

In this paper, a distributed Model Predictive Control (DMPC) is proposed for the secondary voltage and frequency control of islanded microgrid, where each distributed generator (DG) is controlled by a Model Predictive Control (MPC) in the secondary control layer, individually. With considering the nonlinear dynamics of DG with primary control, input‐output feedback schemes are developed for voltage and frequency control separately. Then, all MPCs use the local and neighboring nodes information to solve the optimization problem instead of communicating with a central controller. In this way, the control of the whole system is fully distributed, which allows for a plug‐and‐play. The convergence and stability analysis of the overall closed‐loop system are provided. The simulation result shows the effectiveness of the proposed method.     

Enhanced information reconfiguration for distributed model predictive control for cyber‐physical networked systems

This paper considers a class of cyber‐physical networked systems, which are composed of many interacted subsystems, and are controlled in a distributed framework. The operating point of each subsystem changes with the varying of working conditions or productions, which may cause the change of the interactions among subsystems correspondingly. How to adapt to this change with good closed‐loop optimization performance and appropriate information connections is a problem. To solve this problem, the impaction of a subsystem's control action on the performance of related closed‐loop subsystems is first deduced for measuring the coupling among subsystems. Then, a distributed model predictive control (MPC) for tracking, whose subsystems online reconfigure their information structures, is proposed based on this impaction index. When the operating points changed, each local MPC calculates the impaction indices related to its structural downstream subsystems. If and only if the impaction index exceeds a defined bound, its behavior is considered by its downstream subsystem's MPC. The aim is to improve the optimization performance of entire closed‐loop systems and avoid the unnecessary information connections among local MPCs. Besides, contraction constraints are designed to guarantee that the overall system converges to the set points. The stability analysis is also provided. Simulation results show that the proposed impaction index is reasonable along with the efficiency of the proposed distributed MPC.     

Humanoid walking pattern generation based on model predictive control approximated with basis functions

This paper proposes a real-time walking pattern generator (WPG) based on model predictive control (MPC). Since reducing the calculation time is a crucial problem in real-time WPG, we consider introducing basis functions to reduce the number of control input. The control inputs in the MPC are described by a series of basis functions. Compared with the standard discrete-time MPC formulation, the approach with basis functions requires fewer optimization variables at the cost of decreasing precision. In order to find an appropriate trade-off, two basis functions named Laguerre functions and Haar functions, are tested in this paper. MPC with Laguerre functions decreases more computational load while MPC with Haar functions offers a more accurate solution. The approach is not restricted to Laguerre functions or Haar functions, users can select their own basis functions for different applications and preferences.     

Switched model predictive control for performance enhancement

This article presents a switched model predictive control (MPC) algorithm for non-linear discrete time systems where the weights on the state and control variables in the cost function to be minimised depend on the current value of the state. In so doing, with a reduced computational burden one can easily include, in the problem formulation, a number of control specifications, such as the requirement to avoid critical regions in the state space or to reduce as much as possible the use of some actuators in other zones of the state space. The proposed MPC method has stability properties and is applied for control of a thermal system. The reported simulation results witness its favourable characteristics with respect to a standard MPC implementation.     

基于切换控制的经济模型预测控制

为了提升经济模型预测控制的经济性能指标, 提出一种切换控制策略. 首先, 依据Lyapunov 稳定性理论给出理想和扰动下的两类估计可行域, 并实时检测系统状态; 然后, 根据系统状态所处不同区域, 采用相应的控制器分别实施经济优化、状态驱动和稳态驱动. 所提方法在保证稳定性的同时, 能够为经济性能优化提供更多的在线优化时间和优化自由度, 获得比传统方法更高的经济效益. 通过一个负阻振荡器实例验证了所提出方法的可行性和有效性.

     

高速列车精确停车的鲁棒自触发预测控制

列车精确停车作为列车自动运行(Automatic train operation, ATO)系统的一项核心功能, 对高速列车的安全和高效运行至关重要. 本文针对高速列车停车过程的特点, 考虑在避免控制输出频繁切换的前提下实现高精度的停车曲线跟踪, 提出了基于模型预测控制(Model predictive control, MPC)的精确停车算法. 针对列车停车过程中外部不确定性阻力干扰, 采用鲁棒模型预测控制方法, 提高对外部干扰的鲁棒性. 引入自触发控制策略, 以进一步减少控制输出的频繁切换, 提高停车过程的舒适度. 该方法不需要每个采样时间都求解线性约束二次规划问题, 降低了对系统采样和通信能力的要求, 提高了算法的实用性. 分析结果表明, 高速列车精确停车控制方法的稳定性和性能指标的次优性可以得到保证. 基于高速列车实际运行数据的仿真结果验证了算法的有效性.     

A convex approach for NMPC based on second order Volterra series models

In model predictive control (MPC), the input sequence is computed, minimizing a usually quadratic cost function based on the predicted evolution of the system output. In the case of nonlinear MPC (NMPC), the use of nonlinear prediction models frequently leads to non‐convex optimization problems with several minimums. This paper proposes a new NMPC strategy based on second order Volterra series models where the original performance index is approximated by quadratic functions, which represent a lower bound of the original performance index. Convexity of the approximating quadratic cost functions can be achieved easily by a suitable choice of the weighting of the control increments in the performance index. The approximating cost functions can be globally minimized by convex optimization techniques in order to compute the input sequence. The minimization of the performance index is carried out by an iterative optimization procedure, which guarantees convergence to the solution. Furthermore, for a nominal prediction model, asymptotic stability for the proposed NMPC strategy can be shown. In the case of considering an estimation error in the prediction model, input‐to‐state practical stability is assured. The control performance of the NMPC strategy is illustrated by experimental results. Copyright © 2014 John Wiley & Sons, Ltd.     

双层结构预测控制研究进展

现代工业大系统的优化控制采用递阶结构,其中以预测控制为代表的先进过程控制已经成为重要的一级.目前,主流的工业预测控制技术均采用双层结构,即包含稳态优化层和动态控制层.双层结构预测控制技术可以有效解决复杂工业过程常见的多目标优化、多变量控制的难点问题.本文简要总结了双层结构预测控制的算法,并从控制输入与被控输出稳态关系入手分析了多变量预测控制稳态解的相容性和唯一性,说明了稳态优化的重要性.针对双层结构预测控制与区间预测控制的性能比较、稳态模型的奇异性以及闭环系统动态特性等提出了一些见解,并指出了需要重点研究的主题.     

Reduced order model predictive control for constrained discrete‐time linear systems

A reduced order model predictive control (MPC) is discussed for constrained discrete‐time linear systems. By employing a decomposition method for finite‐horizon linear systems, an MPC law is obtained from a reduced order optimization problem. The decomposition enables us to construct pairs of initial state and control sequence which have large influence on system responses, and it also characterizes the standard LQ control. The MPC law is obtained based on a combination of the LQ control and dominant input sequences over the prediction horizon. The proposed MPC method is illustrated with numerical examples. Copyright © 2011 John Wiley & Sons, Ltd.     

Model predictive control for constrained uncertain piecewise linear systems

For constrained piecewise linear (PWL) systems, the possible existing model uncertainty will bring the difficulties to the design approaches of model predictive control (MPC) based on mixed integer programming (MIP). This paper combines the robust method and hybrid method to design the MPC for PWL systems with structured uncertainty. For the proposed approach, as the system model is known at current time, a free control move is optimized to be the current control input. Meanwhile, the MPC controller uses a sequence of feedback control laws as the future control actions, where each feedback control law in the sequence corresponds to each partitions and the arbitrary switching technique is adopted to tackle all the possible switching. Furthermore, to reduce the online computational burden of MPC, the segmented design procedure is suggested by utilizing the characteristics of the proposed approach. Then, an offline design algorithm is proposed, and the reserved degree of freedom can be online used to optimize the control input with lower computational burden.     

约束非线性系统理想点多目标安全预测控制

考虑具有状态和控制约束的仿射非线性系统多目标安全控制问题,本文提出一种保证安全和稳定的多目标安全模型预测控制(MOSMPC)策略.首先通过理想点逼近方法解决多个控制目标的冲突问题.其次,利用控制李雅普诺夫障碍函数(CLBF)参数化局部控制律,并确定系统不安全域.在此基础上,构造非线性系统的参数化双模控制器,减少在线求解模型预测控制(MPC)优化问题的计算量.进一步,应用双模控制原理和CLBF约束,建立MOSMPC策略的递推可行性和闭环系统的渐近稳定性,并保证闭环系统状态避开不安全域.最后,以加热系统的多目标控制为例,验证了本文策略的有效性.     

Model predictive control of distributed networked control systems with quantization and switching topology

This work is concerned with the robust model predictive control (MPC) for a class of distributed networked control systems (NCSs), in which the input quantization and switching topology are both considered. By utilizing the sector bound approach, the NCSs with quantization are converted into the linear systems with sector bound uncertainties. The topology switching is governed by a switching signal and the dynamic behavior is modeled as a switched control system. A new robust MPC design technique is derived to minimize the upper bound of a weighted quadratic performance index. Moreover, the conditions of both the recursive feasibility of the MPC design and the stability of the resulting closed‐loop system are developed. Finally, simulation results are presented to verify the effectiveness of the proposed MPC design.     

三相脉宽调制整流器的约束预测控制

本文将基于并行神经网络优化的约束模型预测控制(MPC)应用于脉宽调制(PWM)整流器中,提高了电网的质量.在三相静止坐标系下,建立了三相PWM整流器的解耦数学模型,采用约束模型预测控制策略,突破了有限集和无约束条件下预测控制的局限性.为了提高单步优化的速度,采用神经网络优化算法求解模型预测控制的在线优化.在保证系统单位功率因数的前提下,当系统负载突然变化时,具有快速动态响应稳定输出直流电压的性能.采用FPGA控制器实现并行计算,减少了预测控制算法的计算时间.最后,通过仿真和实验结果得到,采用本文的控制策略,总谐波失真(THD)降低了2.5%,达到稳态的时间大约是PI控制算法的五分之一,为12 ms,验证了该方法的可行性和有效性.     

Distributed model predictive control over network information exchange for large-scale systems

A class of large scale systems, which is naturally divided into many smaller interacting subsystems, are usually controlled by a distributed or decentralized control framework. In this paper, a novel distributed model predictive control (MPC) is proposed for improving the performance of entire system. In which each subsystem is controlled by a local MPC and these controllers exchange a reduced set of information with each other by network. The optimization index of each local MPC considers not only the performance of the corresponding subsystem but also that of its neighbours. The proposed architecture guarantees satisfactory performance under strong interactions among subsystems. A stability analysis is presented for the unconstrained distributed MPC and the provided stability results can be employed for tuning the controller. Experiment of the application to accelerated cooling process in a test rig is provided for validating the efficiency of the proposed method.     

Robust MPC for Linear Systems with Structured Time‐Varying Uncertainties and Saturating Actuator

In this paper, a synthesis of model predictive control (MPC) algorithm is presented for uncertain systems subject to structured time‐varying uncertainties and actuator saturation. The system matrices are not exactly known, but are affine functions of a time varying parameter vector. To deal with the nonlinear actuator saturation, a saturated linear feedback control law is expressed into a convex hull of a group of auxiliary linear feedback laws. At each time instant, a state feedback law is designed to ensure the robust stability of the closed‐loop system. The robust MPC controller design problem is formulated into solving a minimization problem of a worst‐case performance index with respect to model uncertainties. The design of controller is then cast into solving a feasibility of linear matrix inequality (LMI) optimization problem. Then, the result is further extended to saturation dependent robust MPC approach by introducing additional variables. A saturation dependent quadratic function is used to reduce the conservatism of controller design. To show the effectiveness, the proposed robust MPC algorithms are applied to a continuous‐time stirred tank reactor (CSTR) process.