Deterministic vs. stochastic performance assessment of iterative learning control for batch processes

The performance assessment of linear time‐invariant batch processes when iterative learning control (ILC) is implemented has been discussed. Previous literatures show that conventional performance assessment cannot be directly applied to batch processes due to the nature of batch operations. Chen and Kong have suggested a new method to assess the control performance of batch processes using optimal ILC as the benchmark. In their work, ILC controllers are assumed to affect either stochastic or deterministic performance but without considering their interaction. This work elaborates the controllers effects on both stochastic and deterministic control performance of batch processes. It is shown that the optimal solution based on the minimum variance control law has a trade‐off between deterministic and stochastic performance, which can be shown by a trade‐off curve. Furthermore, a method is proposed to estimate this curve from routine operating data, against which the performance of ILC controllers can be assessed. Simulation studies are conducted to verify the proposed method. © 2012 American Institute of Chemical Engineers AIChE J, 59: 457–464, 2013     

基于线性时变扰动模型的间歇过程最优迭代学习控制

A batch-to-batch optimal iterative learning control (ILC) strategy for the tracking control of product quality in batch processes is presented. The linear time-varying perturbation (LTVP) model is built for product quality around the nominal trajectories. To address problems of model-plant mismatches, model prediction errors in the previous batch run are added to the model predictions for the current batch run. Then tracking error transition models can be built, and the ILC law with direct error feedback is explicitly obtained. A rigorous theorem is proposed, to prove the convergence of tracking error under ILC. The proposed methodology is illustrated on a typical batch reactor and the results show that the performance of trajectory tracking is gradually improved by the ILC.     

基于输入轨迹参数化的间歇过程迭代学习控制

针对间歇过程的迭代学习控制问题,提出了一种基于输入轨迹参数化的迭代学习控制策略。根据最优输入轨迹的主要形态特征,将其参数化为较少量的决策变量,降低传统迭代学习控制复杂性的同时维持良好的优化控制效果。基于输入轨迹参数化的迭代学习控制策略能保持算法的简洁性和易实现性,在不确定扰动影响下逐步改善产品质量。对一个间歇反应器的仿真研究验证了本文方法的有效性。     

基于广义预测控制的间歇生产迭代优化控制

针对间歇生产,提出了一种基于广义预测控制的批次迭代优化控制策略--BGPC,在间歇过程中引入批次间优化的思想,将迭代学习控制ILC和广义预测控制GPC相结合,在GPC实时结构参数辨识的基础上利用前面批次的模型预测误差修正当前批次的模型预测值.该算法能够有效地克服模型失配、扰动和系统参数变化等情况.文章最后以一个数值例子和间歇反应器为对象进行仿真试验,验证了该算法是有效的.     

Constraint data‐driven optimal terminal ILC of end product quality for a class of I/O constrained batch processes

In order to deal with the I/O constraints in a practical plant, a soft limiter is often added into the control design procedure directly; however, the performance of the soft limiter based control approach will be degraded greatly due to the use of the soft constraints. This paper proposes a data‐driven optimal terminal iterative learning control (constraint‐DDOTILC) approach for the end product quality control of batch processes with I/O hard constraints. To deal with nonlinearities, a novel iterative dynamic linearization method without omitting any information of the original plant is introduced such that the derived linearized data‐driven model is completely equivalent to the original nonlinear system. By transferring all the constraints on the system output, control input, and the change rate of input signals into a linear inequality, a novel constraint‐DDOTILC is developed by minimizing an objective function under the derived linear matrix inequality constraint. The optimal learning gain of the constraint‐DDOTILC can be updated iteratively according to the input and output measurements to enhance the flexibility for modifications and expansions of the controlled plant. Both theoretical analysis and simulation results confirm the effectiveness of the proposed constraint‐DDOTILC.     

一种间歇过程产品质量迭代学习控制策略

针对基于迭代学习控制的间歇过程产品质量优化控制算法难以进行收敛性分析的难题,以数据驱动的神经模糊模型为基础,提出一种新颖间歇过程的产品质量迭代学习控制方法。通过在优化算法中加入了新的约束条件,改变了最优解的搜索空间范围,从而使产品质量在批次轴上收敛,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。     

时变间歇过程的2D-PID自适应控制方法

针对间歇过程存在的参数时变问题,提出一种基于二维PID(2D-PID)迭代学习框架的自适应控制方法。首先,通过粒子群优化算法快速获取初始的2D-PID控制参数。在批次内,采用自调整神经元PID控制器对其进行在线自适应调节。进一步,考虑批次间的重复特性,通过PID型迭代学习控制,以利用历史批次的信息来修正当前批次的调节变量,最终提高控制性能。通过间歇发酵过程的仿真和比较研究,验证了所提出方法的有效性。     

An LMI Method to Robust Iterative Learning Fault-tolerant Guaranteed Cost Control for Batch Processes

Based on an equivalent two-dimensional Fornasini-Marchsini model for a batch process in industry, a closed-loop robust iterative learning fault-tolerant guaranteed cost control scheme is proposed for batch processes with actuator failures. This paper introduces relevant concepts of the fault-tolerant guaranteed cost control and formulates the robust iterative learning reliable guaranteed cost controller （ILRGCC）. A significant advantage is that the proposed ILRGCC design method can be used for on-line optimization against batch-to-batch process uncertainties to realize robust tracking of set-point trajectory in time and batch-to-batch sequences. For the convenience of implementation, only measured output errors of current and previous cycles are used to design a synthetic controller for iterative learning control, consisting of dynamic output feedback plus feed-forward control. The proposed controller can not only guarantee the closed-loop convergency along time and cycle sequences but also satisfy the H∞ performance level and a cost function with upper bounds for all admissible uncertainties and any actuator failures. Sufficient conditions for the controller solution are derived in terms of linear matrix inequalities （LMIs）, and design procedures, which formulate a convex optimization problem with LMI constraints, are presented. An example of injection molding is given to illustrate the effectiveness and advantages of the ILRGCC design approach.     

Iterative learning model predictive control for constrained multivariable control of batch processes

In this paper, we propose a model predictive control (MPC) technique combined with iterative learning control (ILC), called the iterative learning model predictive control (ILMPC), for constrained multivariable control of batch processes. Although the general ILC makes the outputs converge to reference trajectories under model uncertainty, it uses open-loop control within a batch; thus, it cannot reject real-time disturbances. The MPC algorithm shows identical performance for all batches, and it highly depends on model quality because it does not use previous batch information. We integrate the advantages of the two algorithms. The proposed ILMPC formulation is based on general MPC and incorporates an iterative learning function into MPC. Thus, it is easy to handle various issues for which the general MPC is suitable, such as constraints, time-varying systems, disturbances, and stochastic characteristics. Simulation examples are provided to show the effectiveness of the proposed ILMPC.     

Iterative learning control for the systematic design of supersaturation controlled batch cooling crystallisation processes

The paper presents an approach to improve the product quality from batch-to-batch by exploiting the repetitive nature of batch processes to update the operating trajectories using process knowledge obtained from previous runs. The data based methodology is focused on using the linear time varying (LTV) perturbation model in an iterative learning control (ILC) framework to provide a convergent batch-to-batch improvement of the process performance indicator. The major contribution of this work is the development of a novel hierarchical ILC (HILC) scheme for systematic design of the supersaturation controller (SSC) of seeded batch cooling crystallizers. The HILC is used to determine the required supersaturation setpoint for the SSC and the corresponding temperature trajectory required to produce crystals with desired end-point property. The performance and robustness of these approaches are evaluated through simulation case studies. These results demonstrate the potential of the ILC approaches for controlling batch processes without rigorous process models.     

Model predictive control with learning‐type set‐point: Application to artificial pancreatic β‐cell

A novel combination of model predictive control (MPC) and iterative learning control (ILC), referred to learning‐type MPC (L‐MPC), is proposed for closed‐loop control in an artificial pancreatic β‐cell. The main motivation for L‐MPC is the repetitive nature of glucose‐meal‐insulin dynamics over a 24‐h period. L‐MPC learns from an individual's lifestyle, inducing the control performance to improve from day to day. The proposed method is first tested on the Adult Average subject presented in the UVa/Padova diabetes simulator. After 20 days, the blood glucose concentrations can be kept within 68–145 mg/dl when the meals are repetitive. L‐MPC can produce superior control performance compared with that achieved under MPC. In addition, L‐MPC is robust to random variations in meal sizes within ±75% of the nominal value or meal timings within ±60 min. Furthermore, the robustness of L‐MPC to subject variability is validated on Adults 1–10 in the UVa/Padova simulator. © 2009 American Institute of Chemical Engineers AIChE J, 2010     

一类连续/间歇过程的迭代学习模型预测控制

An iterative learning model predictive control (ILMPC) technique is applied to a class of continuous/batch processes. Such processes are characterized by the operations of batch processes generating periodic strong disturbances to the continuous processes and traditional regulatory controllers are unable to eliminate these periodic disturbances. ILMPC integrates the feature of iterative learning control (ILC) handling repetitive signal and the flexibility of model predictive control (MPC). By on-line monitoring the operation status of batch processes, an event-driven iterative learning algorithm for batch repetitive disturbances is initiated and the soft constraints are adjusted timely as the feasible region is away from the desired operating zone. The results of an industrial appli-cation show that the proposed ILMPC method is effective for a class of continuous/batch processes.     

Using process data to generate an optimal control policy via apprenticeship and reinforcement learning

Reinforcement learning (RL) is a data-driven approach to synthesizing an optimal control policy. A barrier to wide implementation of RL-based controllers is its data-hungry nature during online training and its inability to extract useful information from human operator and historical process operation data. Here, we present a two-step framework to resolve this challenge. First, we employ apprenticeship learning via inverse RL to analyze historical process data for synchronous identification of a reward function and parameterization of the control policy. This is conducted offline. Second, the parameterization is improved online efficiently under the ongoing process via RL within only a few iterations. Significant advantages of this framework include to allow for the hot-start of RL algorithms for process optimal control, and robust abstraction of existing controllers and control knowledge from data. The framework is demonstrated on three case studies, showing its potential for chemical process control.     

Optimization of nonlinear process based on sequential extreme learning machine

In this paper, a new approach to the optimal control with constraints is proposed to achieve a desired end product quality for nonlinear processes based on new kernel extreme learning machine (KELM). The contributions of the paper are as follows: (1) In existing ILC algorithm, the model was built only between manipulated input variables U and output variables Y without considering the state variables. However, the states variables X_state are important in the industrial processes, which are usually constrained. In this paper, the variables are divided into state variables X_state, manipulated input variables U and output Y in the process of modeling. Then ΔU can be obtained by batch-to-batch iterative learning control separately. Kernel algorithm is added to ELM. (2) Constraints of state variables X_state and the input variables U are considered in the current version. PSO is used to solve the optimization problem. (3) Kernel trick is introduced to improve accuracy of ELM modeling. New KELM algorithm is proposed in the current version. The input trajectory for the next batch is accommodated by searching for the optimal value through the error feedback at a minimum cost. The particle swarm optimization algorithm is used to search for the optimal value based on the iterative learning control (ILC). The proposed approach has been shown to be effective and feasible by applying bulk polymerization of the styrene batch process and fused magnesium furnace.     

Design and Analysis of Integrated Predictive Iterative Learning Control for Batch Process Based on Two-dimensional System Theory

Based on the two-dimensional (2D) systemtheory, an integrated predictive iterative learning control (2D-IPILC) strategy for batch processes is presented. First, the output response and the error transition model predictions along the batch index can be calculated analytically due to the 2D Roesser model of the batch process. Then, an integrated framework of combining iterative learning control (ILC) andmodel predictive control(MPC) is formed reasonably. The output of feedforward ILC is estimated on the basis of the predefined process 2D model. By minimizing a quadratic objective function, the feedback MPC is introduced to obtain better control performance for tracking problem of batch processes. Simulations on a typical batch reactor demonstrate that the satisfactory tracking performance as well as faster convergence speed can be achieved than traditional proportion type (Ptype) ILC despite the model error and disturbances.     

Delay‐range‐dependent guaranteed cost control for batch processes with state delay

A guaranteed cost control scheme is proposed for batch processes described by a two‐dimensional (2‐D) system with uncertainties and interval time‐varying delay. First, a 2‐D controller, which includes a robust feedback control to ensure performances over time and an iterative learning control to improve the tracking performance from cycle to cycle, is formulated. The guaranteed cost law concept of the proposed 2‐D controller is then introduced. Subsequently, by introducing the Lyapunov–Krasovskii function and adding a differential inequality to the Lyapunov function for the 2‐D system, sufficient conditions for the existence of the robust guaranteed cost controller are derived in terms of matrix inequalities. A design procedure for the controller is also presented. Furthermore, a convex optimization problem with linear matrix inequality (LMI) constraints is formulated to design the optimal guaranteed cost controller that minimizes the upper bound of the closed‐loop system cost. The proposed control law can stabilize the closed‐loop system as well as guarantee H_∞ performance level and a cost function with upper bounds for all admissible uncertainties. The results can be easily extended to the constant delay case. Finally, an illustrative example is given to demonstrate the effectiveness and advantages of the proposed 2‐D design approach. © 2013 American Institute of Chemical Engineers AIChE J, 59: 2033–2045, 2013     

Data-driven modelling methods in sintering process: Current research status and perspectives

The sintering process, as a primary modus of the blast furnace ironmaking industry, has enormous economic value and environmental protection significance for the iron and steel enterprises. Recently, with the emergence of artificial intelligence and big data, data-driven modelling methods in the sintering process have increasingly received the researchers' attention. But now, there is still no systematic review of the data-driven modelling approaches in the sintering process. Therefore, in this article, we conduct a comprehensive overview and prospects on the data-driven models for the purpose of intelligent sintering. First, the mechanism and characteristics of the sintering process are introduced and analyzed elaborately. Second, the detailed research status of the sintering process is illustrated from four aspects: key parameters prediction, control, optimization, and others. Finally, several challenges and promising modelling methods such as deep learning in the sintering process are outlined and discussed for future research.     

Investigation of a Model-Based Deep Reinforcement Learning Controller Applied to an Air Separation Unit in a Production Environment

The need for load flexibility and increased efficiency of energy-intensive processes has become more and more important in recent years. Control of the process variables plays a decisive role in maximizing the efficiency of a plant. The widely used control models of linear model predictive controllers (LMPC) are only partly suitable for nonlinear processes. One possibility for improvement is machine learning. In this work, one approach for a purely data-driven controller based on reinforcement learning is explored at an air separation plant (ASU) in productive use. The approach combines the model predictive controller with a data-generated nonlinear control model. The resulting controller and its control performance are examined in more detail on an ASU in real operation and compared with the previous LMPC solution. During the tests, stable behavior of the new control concept could be observed for several weeks in productive operation.