化工过程约束优化控制的可行性分析及约束处理

在化工领域过程控制中,普遍存在着各种对输出变量、输入变量甚至中间变量的约束。不同约束条件之间的矛盾会造成约束条件无法全部满足,优化控制器无可行解,给实际生产造成负面影响。从凸体几何角度,将化工生产过程中约束优化控制的可行性判定转化为凸多面体是否相交的问题,将不可行时合理的约束处理方案转化为一系列线性规划或非线性规划问题,提出无需人为参与的自动进行约束优化控制可行性分析和约束调整的算法。Shell公司提供的重油分馏塔典型案例实验证明,该算法能够在约束优化控制不可行时自动有效地进行合理的约束调整,超调量小,控制作用变化平缓,且有一定控制裕量。     

The synthesis of nearly optimal on-off controllers with application to chemical processes

A new technique is presented for designing a set of on-off controllers for processes which can be approximated by linear-lumped parameter models of any order. The design is based on the minimization of a quadratic function of the process outputs. Each higher member of the set of systems approaches optimal control ever more closely. Our major concern, however, is with the lowest member of the set which is implemented by using only a relay and a simple summing device. Several analog computer examples are presented using this control scheme. In all cases, the system response is excellent from both quantitative and qualitative viewpoints.

One of the examples presented is the control of a simulated six stage liquid—liquid extraction unit. Here, the lowest level control is shown to outperform an experimentally tuned standard three-mode type of control as well as a control strategy based on dynamic programming which was developed for the same process by a previous investigator.     

Optimal design of chemical processes with chance constraints

Generally, chemical processes (CP) are designed with the use of inaccurate mathematical models. Therefore, it is important to create a chemical process that guarantees satisfaction of all design specifications either exactly or with some probability. The paper considers the issue of chemical process optimization when at the operation stage the design specification should be met with some probability and the control variables can be changed. We have developed a common approach for solving the broad class of optimization problems with normally distributed uncertain parameters. This class includes the one-stage and two-stage optimization problems with chance constraints. This approach is based on approximate transformation of chance constraints into deterministic ones.     

Explicit distributed model predictive control design for chemical processes under constraints and uncertainty

In multivariable industrial processes, the common distributed model predictive control strategy is usually unable to deal with complex large-scale systems efficiently, especially under system constraints and high control performance requirements. Based on this situation, we use the distributed idea to divide the large-scale system into multiple subsystems and transform them into the state space form. Combined with the output tracking error term, we build an extended non-minimal state space model that includes output error and measured output and input. When dealing with system constraints, the new constraint matrix is divided into range and kernel space by using the explicit model predictive control algorithm, which reduces the difficulty of solving constraints in the extended system and further improves the overall control performance of the system. Finally, taking the coke furnace pressure control system as an example, the proposed algorithm is compared with the conventional distributed model predictive control algorithm using non-minimal state space, and the simulation results show the feasibility and superiority of this method.     

A unified approach to the design of advanced proportional-integral-derivative controllers for time-delay processes

A unified approach for the design of proportional-integral-derivative (PID) controllers cascaded with first-order lead-lag filters is proposed for various time-delay processes. The proposed controller’s tuning rules are directly derived using the Padé approximation on the basis of internal model control (IMC) for enhanced stability against disturbances. A two-degrees-of-freedom (2DOF) control scheme is employed to cope with both regulatory and servo problems. Simulation is conducted for a broad range of stable, integrating, and unstable processes with time delays. Each simulated controller is tuned to have the same degree of robustness in terms of maximum sensitivity (Ms). The results demonstrate that the proposed controller provides superior disturbance rejection and set-point tracking when compared with recently published PID-type controllers. Controllers’ robustness is investigated through the simultaneous introduction of perturbation uncertainties to all process parameters to obtain worst-case process-model mismatch. The process-model mismatch simulation results demonstrate that the proposed method consistently affords superior robustness.     

基于无约束迭代学习的间歇生产过程优化控制

针对基于迭代学习控制的间歇过程优化控制算法难以进行收敛性分析的难题,本文基于数据驱动的神经模糊模型提出一种新颖的间歇过程无约束迭代学习控制方法,通过调节因子的变化去除了约束条件,使控制轨迹在批次轴上收敛,并创新性地对优化问题的收敛性给出了严格的数学证明。在理论研究的基础上,将本文提出的算法用于间歇连续反应釜的终点质量控制研究,仿真结果验证了本文算法的有效性和实用价值,为间歇过程的优化控制提供了一条新途径。     

定性/定量集成优化控制器的形式化设计方法

在化工过程中广泛存在着混杂系统的特性,即系统的信息结构由系统事件和系统状态构成。本文以时段演算为工具提出面向混合信息结构的混杂系统在定性/定量双重指标约束下的形式化设计方法,它将时段演算应用于混杂系统的需求刻划,得出控制系统的定性/定量行为和语义的定义,给出控制系统的行为描述形式化、优化描述形式化和基于时段演算的形式化设计步骤。最后结合一个应用实例,说明设计方法的有效性。     

氧化铝蒸发过程的混合智能优化设定控制

氧化铝蒸发过程的关键工艺指标碱液浓度不能在线检测且与控制回路输出之间的动态特性难以用精确的数学模型描述,采用已有的优化控制方法不能实现上述运行层的优化。针对上述问题提出了由基于案例推理的预设定模型,基于专家规则的前馈、反馈补偿模型以及基于块式偏最小二乘的软测量模型组成的混合智能优化设定控制方法。采用某氧化铝厂蒸发过程的实际数据进行仿真实验,实验研究表明该方法可以有效地将碱液浓度控制在工艺要求的区间内。     

Model predictive control for multivariable unstable processes with constraints on manipulated variables

The original MPC(Model Predictive Control) algorithm cannot be applied to open loop unstable systems, because the step responses of the open loop unstable system never reach steady states. So when we apply MPC to the open loop unstable systems, first we have to stabilize them by state feedback or output feedback. Then the stabilized systems can be controlled by MPC. But problems such as valve saturation may occur because the manipulated input is the summation of the state feedback output and the MPC output. Therefore, we propose Quadratic Dynamic Matrix Control(QDMC) combined with state feedback as a new method to handle the constraints on manipulated variables for multivariable unstable processes. We applied this control method to a single-input-single-output unstable nonlinear system and a multi-input-multi-output unstable system. The results show that this method is robust and can handle the input constraints explicitly and also its control performance is better than that of others such as well tuned PI control. Linear Quadratic Regulator (LQR) with integral action.     

Model predictive control for multivariable unstable processes with constraints on manipulated variables

The original MPC(Model Predictive Control) algorithm cannot be applied to open loop unstable systems, because the step responses of the open loop unstable system never reach steadystates. So when we apply MPC to the open loop unstable systems, first we have to stabilize them by state feedback or output feedback. Then the stabilized systems can be controlled by MPC. But problems such as valve saturation may occur because the manipulated input is the summation of the state feedback output and the MPC output. Therefore, we propose Quadratic Dynamic Matrix Control(QDMC) combined with state feedback as a new method to handle the constraints on manipulated variables for multivariable unstable processes. We applied this control method to a single-input-single-output unstable nonlinear system and a multi-input-multi-output unstable system. The results show that this method is robust and can handle the input constraints explicitly and also its control performance is better than that of others such as well tuned PI control. Linear Quadratic Regulator (LQR) with integral action.     

Distributed economic model predictive control for operational safety of nonlinear processes

Achieving operational safety of chemical processes while operating them in an economically‐optimal manner is a matter of great importance. Our recent work integrated process safety with process control by incorporating safety‐based constraints within model predictive control (MPC) design; however, the safety‐based MPC was developed with a centralized architecture, with the result that computation time limitations within a sampling period may reduce the effectiveness of such a controller design for promoting process safety. To address this potential practical limitation of the safety‐based control design, in this work, we propose the integration of a distributed model predictive control architecture with Lyapunov‐based economic model predictive control (LEMPC) formulated with safety‐based constraints. We consider both iterative and sequential distributed control architectures, and the partitioning of inputs between the various optimization problems in the distributed structure based on their impact on process operational safety. Moreover, sufficient conditions that ensure feasibility and closed‐loop stability of the iterative and sequential safety distributed LEMPC designs are given. A comparison between the proposed safety distributed EMPC controllers and the safety centralized EMPC is demonstrated via a chemical process example. © 2017 American Institute of Chemical Engineers AIChE J, 63: 3404–3418, 2017     

Safe model-based reinforcement learning for nonlinear optimal control with state and input constraints

Safety is a critical factor in reinforcement learning (RL) in chemical processes. In our previous work, we had proposed a new stability-guaranteed RL for unconstrained nonlinear control-affine systems. In the approximate policy iteration algorithm, a Lyapunov neural network (LNN) was updated while being restricted to the control Lyapunov function, and a policy was updated using a variation of Sontag's formula. In this study, we additionally consider state and input constraints by introducing a barrier function, and we extend the applicable type to general nonlinear systems. We augment the constraints into the objective function and use the LNN added with a Lyapunov barrier function to approximate the augmented value function. Sontag's formula input with this approximate function brings the states into its lower level set, thereby guaranteeing the constraints satisfaction and stability. We prove the practical asymptotic stability and forward invariance. The effectiveness is validated using four tank system simulations.     

Analytical design of PID controller cascaded with a lead-lag filter for time-delay processes

An analytical method for the design of a proportional-integral-derivative (PID) controller cascaded with a second-order lead-lag filter is proposed for various types of time-delay process. The proposed design method is based on the IMC-PID method to obtain a desired, closed-loop response. The process dead time is approximated by using the appropriate Pade expansion to convert the ideal feedback controller to the proposed PID·filter structure with little loss of accuracy. The resulting PID·filter controller efficiently compensates for the dominant process poles and zeros and significantly improves the closed-loop performance. The simulation results demonstrate the superior performance of the proposed PID·filter controller over the conventional PID controllers. A guideline for the closed-loop time constant, λ, is also suggested for the FOPDT and SOPDT models.     

Hybrid simulation-optimization based approach for the optimal design of single-product biotechnological processes

In this work, we present a systematic method for the optimal development of bioprocesses that relies on the combined use of simulation packages and optimization tools. One of the main advantages of our method is that it allows for the simultaneous optimization of all the individual components of a bioprocess, including the main upstream and downstream units. The design task is mathematically formulated as a mixed-integer dynamic optimization (MIDO) problem, which is solved by a decomposition method that iterates between primal and master sub-problems. The primal dynamic optimization problem optimizes the operating conditions, bioreactor kinetics and equipment sizes, whereas the master levels entails the solution of a tailored mixed-integer linear programming (MILP) model that decides on the values of the integer variables (i.e., number of equipments in parallel and topological decisions). The dynamic optimization primal sub-problems are solved via a sequential approach that integrates the process simulator SuperPro Designer^® with an external NLP solver implemented in Matlab^®. The capabilities of the proposed methodology are illustrated through its application to a typical fermentation process and to the production of the amino acid L-lysine.     

Computer-aided multivariable control system design for processes with time delays

A computer-aided control system design package is described which handles processes with multiple time delays and allows subsequent design of interaction compensators. The package synthesizes control system designs for complex problems in a short time while also providing dynamic simulation for the purpose of evaluating controller performance. Example problems are presented along with a case study on sensitivity to errors in the delays.     

Approximation of the optimal minimum-phase output for control of nonlinear nonminimum-phase processes

Nonminimum-phase parts are better removed in the feedback loop like the time delay term. For this, Wright and Kravaris [1992] proposed the concept of optimal minimum-phase output to control nonlinear nonminimumphase processes. However, their optimal minimum-phase output has no analytic solutions for processes with more than three state variables. Here, methods for analytic minimum-phase outputs approximating the optimal ones are proposed, having no limitations in the number of state variables. The proposed methods provide analytic solutions for processes with three state variables and simple numerical solutions for those with more state variables.     

Decentralized fault-tolerant control system design for unstable processes

Fault-tolerant control is an important issue in control of mission critical processes. In this paper, a new approach to fault-tolerant control of unstable processes is proposed based on the Passivity Theorem. The control system is designed in two sequential steps: A multi-loop proportional controller is used to stabilize the unstable process; a passivity-based decentralized unconditionally stabilizing (DUS) controller is then applied to the stabilized process. While the multi-loop stabilizing controllers need to be built with redundancy, the DUS controller is inherently fault tolerant and can maintain closed-loop stability when any of its loops fail. By using a stabilizing proportional controller with the fewest loops, control redundancy can be reduced to the minimum level.     

Plantwide control for optimal operation of industrial-scale crude distillation unit processes

In this article, a combination of the wavelet neural network framework and the line-up competition algorithm is used to solve the economic optimization algorithm for an industrial-scale atmospheric distillation column (ADC) process. Compared to the relevant measuring data from Sinopec Wuhan Petroleum Group Company, China, the first optimal operating conditions show that the increments of the duties of furnace and pump-arounds of the ADC can effectively improve oil production. In our approach, the preflash column (PFC) coupled with ADC is denoted as an industrial-scale crude distillation unit (CDU) process. Since the PFC can produce light naphtha and reduce the furnace duty and steam consumption of ADC, it is verified that the CDU process provides the higher economic potential than ADC. Based on the second optimal operating conditions, the plantwide control strategy is employed to operate the system safely as well as regulate the outputs of the plant in the presence of inlet perturbations. Within the plantwide control framework, the inventory control aims to keep the controlled variables close to the desired operating condition and the quality control loops use a combination of inferential predictions and feedforward ratios to effectively suppress the temperature spikes of trays and furnaces. Finally, the simulations show that the product quality is guaranteed due to no offset ASTM D86 distillation temperature responses.