Research on Theory and Applications of Model Predictive Control for Complex Industrial Processes

Zhao Jun,Xu Zuhua,Qian Jixin

DOI: https://doi.org/10.3969/j.issn.1674-2605.2009.03.001

2009-01-01

Abstract:The paper takes a comprehensive and deep insight in model predictive control algorithm, and presents some renovation points after summarizing a lot of previous research reports in these fields.

What problem does this paper attempt to address?

Process control: Art or practice

Jian Chu,Hongye Su,Feng Gao,Jun Wu

DOI: https://doi.org/10.1016/s1474-6670(17)45150-0

IF: 9.4

1998-01-01

Annual Reviews in Control

Abstract:This paper addresses the development of industrial process control. The functions of the sensors, controllers and actuators in industrial processes are analysed. Then, from application viewpoint, we discussed the progress of process control algorithms from PID to advanced algorithms. Consequently, the distance from modern control theory to the application in industrial process is studied in depth. At last, a case study and the perspectives of industrial process control are given.
Predictive Functional Control for Multivariable First Order plus Dead Time System

NING Cui,ZHANG Quan-ling,SU Hong-ye

DOI: https://doi.org/10.3969/j.issn.1002-0411.2007.06.009

2007-01-01

Information and Control

Abstract:A multivariable predictive functional control algorithm is presented,which is a branch of model predictive techniques family.It is deduced with the theory of predictive functional control and Smith predictor,and is the extension of single variable predictive functional control.A computer simulation case study for the control of Shell Oil's heavy oil fractionator is presented.The result displays that the presented multivariable predictive functional control algorithm is of better control performance than the single variable predictive functional control in the multivariable first order plus dead time system.
An Improved Model Predictive Control Algorithm Based on Multi-Degree-of-Freedom Performance Index

徐祖华,赵均,钱积新

DOI: https://doi.org/10.3321/j.issn:0372-2112.2008.05.013

2008-01-01

Abstract:Traditional model predictive control is a one-degree-of-freedom controller,so good dynamic response performance of both set-point tracking and disturbance rejection characteristics can't be achieved simultaneously.An improved model predictive control algorithm based on multi-degree-of-freedom performance index is proposed for the industrial plant to achieve good dynamic response performance of set-point tracking and disturbance rejection and model mismatch.The simulated results are given to demonstrate the effectiveness of the designed controller.
Studies and Applications of Predictive Functional Control in Temperature Tracking of Chemical Reactors

张泉灵,王树青

DOI: https://doi.org/10.3969/j.issn.1000-8152.2001.04.019

2001-01-01

Abstract:An advanced control strategy as the combination of predictive functional control and PID control is studied. With an industrial application background of temperature control in chemical reactor, predictive functional controller and transparent control scheme are designed. Simulation results show that the transparent control system obtains good tracking control performance, enough disturbance rejection and good robustness. At the same time, the advanced control strategy has been implemented in the SUPCON JX-300 distributed control systems and has provided good results.
Key Technology and Industrial Application of Model Predictive Control Engineering Software

赵均,李田鹏,钱积新

DOI: https://doi.org/10.3969/j.issn.1671-5896.2004.04.012

2004-01-01

Abstract:Key technology of steady-state optimization and dynamic control in MPC(Model Predictive Control) engineering software are proposed, based on our self-development MPC software—Frontapc. The priority strategy to handle multi-objective optimization applied in steady-state optimization and the method to achieving the unique optimization steady state are proposed. Zone control strategy cooperated with funnel constrains is presented for the purpose to improve the robust control quality. The efficiency of this MPC software is illustrated by the successful application in distillation control system of acetaldehyde production process.
Dynamic Modeling And Nonlinear Predictive Control Based On Partitioned Model And Nonlinear Optimization

Ridong Zhang,Anke Xue,Shuqing Wang

DOI: https://doi.org/10.1021/ie102211x

2011-01-01

Industrial & Engineering Chemistry Research

Abstract:The paper presents a combination modeling procedure and the implementation of a nonlinear predictive control scheme for the optimization of industrial chemical processes. The model structure is first based on a simple step response method. This provides a way to use prior knowledge about the dynamics, which has a general validity, while additional information about the process behavior is derived from measured plant data-model error. This data error driven model framework is applicable for a wide range of chemical operating units under a certain control policy. The same idea is also used to solve the online optimization problem in the predictive controller. The efficiency and effectiveness of the modeling training algorithm and the nonlinear predictive control approach are demonstrated through a coke furnace case study. A good model fitting for the nonlinear plant is obtained by using the new method. A comparison with traditional approaches shows that the new algorithm can considerably reduce modeling error and improve control accuracy.
Get the design right, the rest is automatic: Towards an adaptive MPC controller

Yucai Zhu,Zuhua Xu,Jun Zhao,Kai Han,Jixin Qian,Weixin Li

2009-01-01

Abstract:Model predictive control technology has brought huge benefit in the refining/petrochemical industry and starts to attract attention of other process industries. However, sustainability is a big challenge in MPC applications due to its technical difficulties and high cost in human resources. In order to considerably reduce the difficulties and costs of MPC deployment and maintenance, we introduce an adaptive MPC controller. The adaptive MPC consists of three modules: an MPC control module, an online identification module and a control monitor module. The three modules work together coherently in real-time. During MPC commissioning, the online identification module and the MPC control module work together and perform various steps in MPC implementation automatically. When the MPC controller is online, the control monitor module continuously monitors the MPC performance and model quality. When control performance degradation and considerable model error are detected, monitor module will start the maintenance by activating the online identification module. The identification module will re-identify the model and replace the old model. Most of the steps can be done automatically under the supervision of the operator. The scarce control expert is only needed in controller design. A prototype of the adaptive MPC controller has applied successfully to two PTA units and the results will be reported.
The Situation and Prospect of Predictive Control

HUANG Xiang-yun,ZHU Xue-feng

DOI: https://doi.org/10.3969/j.issn.1007-7324.2005.02.008

2005-01-01

Abstract:Nowadays the predictive control is widely used in industrial process,and the theoretical analysis has got much development.The situation and industrial application of different predictive control methods are analyzed;the problems and progressive tendency are also discussed.
Model-based Predictive Control for Industrial Process-A Survey

LI Shao-yuan

DOI: https://doi.org/10.14107/j.cnki.kzgc.2010.04.010

2010-01-01

Abstract:Model-based predictive control has made a significant impact on academic control area and industrial application since it was proposed in 1970's.Especially in industrial process control,predictive control has been considered to be an important advanced control method.According to the characteristics of industrial process control and further demand,the survey about the new advances in predictive control for industrial process control area in recent years is presented,including;real-time idenfifioation method for the idccstrial process plant,data-driven predictive control,netowrked predictive contnol.Moreover,the development in the future is analyzed,the european FP7 STREP project:Hierarchical and distributed model predictive control of large-scale systems is introduced.
An Engineering-Effective Adaptive Model Algorithm Control (AMAC) and Simulation

Chen Qi,Liu Meng,Wang Shu-qing,Wang Ji-cheng

DOI: https://doi.org/10.1016/s1474-6670(17)53344-3

1989-01-01

IFAC Proceedings Volumes

Abstract:There has been extensive interest in model algorithm control (MAC) and its application to industrial processes for dealing with time-varying parameters, uncertain characteristics. But the control based on basic MAC method still exists some problems. One of the problems is how to design MAC system to improve the robustness to prior choice of model. The second is how to use the adaptive algorithm for on-line estimation which doesn't need a huge amount of computational time that influence the regulation performance for engineering application. In this paper, a novel adaptive model algorithm control (AMAC) is developed with consideration for dealing with the model prediction error and bounded disturbances and noise, combined with the auto-stop parameter estimation technique. Based on this method, the complex system can be described with low-term system, which results in reducing computational time, and improving the robustness to model structure for effective control of complex chemical processes. Finally, according to the proposed AMAC method, two typical processes are studied by computer simulation. The results show this AMAC method is much advantaged with robustness to modelling error and different bounded disturbance and can be easily applied to control the complex processes.
Applications Survey of Predictive Control in Process Control

Zhang Chunfeng,Zou Xinjie,Yu Zhangguo

2006-01-01

Abstract:Nowadays, model predictive control (MPC) is widely used in industrial processes, and the theoretical analysis has got much development. In this paper, the methods and characters of MPC are introduced, and the applications of MPC in process control field are analyzed. Meanwhile, the existing problems and progressive tendency are also discussed.
Recent Advances on Predictive Control

Zhu Jing

2001-01-01

Abstract:With the widely application of predictive control in industrial process,the theoretical research on predictive control has made rapid progress recently.In this paper,we survey the predictive control on theoretical and application. Furthermore,we mainly analyze stability and robustness of MPC.At last,we put forward some advices on the development of the MPC.
Recent Advances in Research on Predictive Control Performance

XI Yugeng,GENG Xiaojun,CHEN Hong

DOI: https://doi.org/10.3969/j.issn.1000-8152.2000.04.001

2000-01-01

Abstract:With the wide application of predictive control in industrial processes, the theoretical research on predictive control also achieves great development in recent years. In this paper, recent advance in research on predictive control performance is surveyed, which includes the stability, robustness, feasibility of predictive control systems, as well as nonlinear predictive control.
Analysis and Synthesis of Predictive Control for a Class of Industrial Processes

Li, Junyi,Xi, Yugeng

DOI: https://doi.org/10.23919/acc.1992.4792568

1992-01-01

Abstract:Predictive control has found wide applications in process control field, but its theoretical analysis and synthesis are still lacking. In this paper the predictive control system for a class of industrial processes is investigated. The performance indices of the controlled system are quantitatively analyzed. A synthesis procedure for predictive control systems without simulation trials is also proposed.
Design Of Model Predictive Controller Based On Neural Systems

Ge Ming,Xu Zhe,Sun Youxian

1997-01-01

Abstract:A neural networks-based model predictive (NNMP) control strategy is presented for control of complex chemical processes. In this control strategy, the neural networks are combined with the model predictive controller and trained on-line by minimizing the controller output corresponding to the nonmodelled effect. The NNMP controller is applied to some chemical processes and compared with the conventional model predictive (CMP) controller. The results show that the NNMP controller is more effective than the CMP controller, especially in the control of complex processes with time delays, nonlinearity, and so on. The NNMP controller also has the ability, because of its learning capability, to overcome the disadvantages caused by model mismatch.
Overview of modeling algorithm in predictive control

Hongli Chen,Guocheng Zhang,Lei Wan

2011-01-01

Abstract:In recent years, predictive control has been applied in many large industrial control areas as an effective control method. However, the key problem is how to establish a predictive model approaching real system with high precision. According to the developmental character of the industrial control system, the modeling problems related to predictive control in industrial control areas for years were presented in this paper, including the methods and theories. Furthermore, the characters, applicable conditions and relative merits were summarized. It is also clarified that, in the future, the research orientation and the development of the modeling problem in predictive control are predicted and expected.
Research on the Intelligent Predictive Control of Product Quality

乐清洪,滕霖,王润孝,朱名铨

DOI: https://doi.org/10.3969/j.issn.1005-2402.2004.11.017

2004-01-01

Abstract:How to implement the automation and intelligence of process quality control (PQC) has been an important researching issue in the modern manufacturing system. In this paper, the application status of quality predictive control is introduced. An intelligent predictive control system model based on time series of quality characteristics is proposed to use for PQC. Based on this system, a dynamic neural network is used to track the change of quality characteristics during manufacturing process. Numerical simulation and practical experiment show good results.
Fuzzy-precise combined control system

Ping Li,Hui Wang,Jixin Qian,Youxian Sun,Chunhui Zhou

1994-01-01

Kongzhi Lilun Yu Yinyong/Control Theory and Applications

Abstract:The existing control theory is divided into two parts in this paper, i.e. precise control theory and fuzzy control theory. After investigating the advantages and disadvantages of the precise control and fuzzy control respectively, the so-called fuzzy-precise combined control strategy is proposed from the view of practical application in the paper. This new strategy will bring maximum control benefits with minimum design efforts. As the application background, the fuzzy-precise combined control strategy has applied into four full scale paper machine control systems, and all of them got good control results.
Computer Modeling, Simulation and Control of Complex Industrial Process:A Survey

卢荣德,陈宗海,王雷

DOI: https://doi.org/10.3321/j.issn:1001-506x.2002.01.015

2002-01-01

Abstract:Complex industrial process computer control is a frontier of information and control science. It is due to the complexity of its background that few substantial progresses have been made yet in practical applications, even on the theoretical basis.Therefore,it is a wise choice to realize intelligent control step by step for large-scale complex industrial process. This paper tries to probe into the inherent essence and deep-rooted causes of the process in the scope of complex industrial process computer modeling,simulation and control, and also presents its natural limits, with the purpose to contribute this to the people who are engaged in the specific scientific research.;
Predictive Control for Distributed Systems-a Survey

李少远

DOI: https://doi.org/10.3969/j.issn.1003-0492.2013.z1.008

2013-01-01

Abstract:Model-based predictive control has made a significant impact on academic control area and industrial applications since it was proposed in 1970’s. Especially in industrial process control, with the development of the intelligent sensor, actuator and controller implementation, the control system structure has been employed in distributed mode from the centralized mode, so this survey about the new advances in predictive control for industrial process control area in recent years is presented in this paper. Moreover, the development in the future is analyzed, the European FP7 STREP project:Hierarchical and Distributed Model Predictive Control of Large-Scale Systems is introduced.

Research on Theory and Applications of Model Predictive Control for Complex Industrial Processes

Process control: Art or practice

Predictive Functional Control for Multivariable First Order plus Dead Time System

An Improved Model Predictive Control Algorithm Based on Multi-Degree-of-Freedom Performance Index

Studies and Applications of Predictive Functional Control in Temperature Tracking of Chemical Reactors

Key Technology and Industrial Application of Model Predictive Control Engineering Software

Dynamic Modeling And Nonlinear Predictive Control Based On Partitioned Model And Nonlinear Optimization

Get the design right, the rest is automatic: Towards an adaptive MPC controller

The Situation and Prospect of Predictive Control

Model-based Predictive Control for Industrial Process-A Survey

An Engineering-Effective Adaptive Model Algorithm Control (AMAC) and Simulation

Applications Survey of Predictive Control in Process Control

Recent Advances on Predictive Control

Recent Advances in Research on Predictive Control Performance

Analysis and Synthesis of Predictive Control for a Class of Industrial Processes

Design Of Model Predictive Controller Based On Neural Systems

Overview of modeling algorithm in predictive control

Research on the Intelligent Predictive Control of Product Quality

Fuzzy-precise combined control system

Computer Modeling, Simulation and Control of Complex Industrial Process:A Survey

Predictive Control for Distributed Systems-a Survey