Robust iterative learning control of an exothermic semi-batch chemical reactor

Mezghani, M.; Roux, G.; Cabssud, M.; Dahhou, B.; Lann, M.V. Le; Casamatta, G.

doi:10.1016/s0378-4754(01)00317-2

Cited by 22 publications

(25 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It was originally proposed for robot manipulator (Arimoto et al, 1984) and has been studied for many industrial processes (Bristow et al, 2006;Ahn et al, 2007). In the field of process control, many studies have been conducted for chemical batch reactor (Mezghani et al, 2001;Lee and Lee, 2003) and semiconductor processes (Choi and Do, 2001;Yu et al, 2014). General ILC can achieve exponential or monotonic convergence along the batch index; however, it cannot reject real-time disturbances because ILC is basically open-loop control within a batch.…”

Section: Introductionmentioning

confidence: 99%

Iterative Learning Control Integrated with Model Predictive Control for Real-Time Disturbance Rejection of Batch Processes

Lee

2017

J. Chem. Eng. Japan / JCEJ

View full text Add to dashboard Cite

In the present paper, iterative learning control (ILC) is integrated with a model predictive control (MPC) technique to reject real-time disturbances. The proposed scheme is called iterative learning model predictive control (ILMPC). ILC is an e ective control technique for batch processes, but it is not a real-time feedback controller. Thus, it should be combined with MPC for real-time disturbance rejection. The existing ILMPC techniques make the error converge to zero. However, if the error converges to zero, an impractical input trajectory may be calculated. We use a generalized objective function to independently tune weighting factors of manipulated variable change with respect to both the time index and batch horizons. If the generalized objective function is used, output error converges to non-zero values. We provide convergence analysis for both cases of zero convergence and non-zero convergence.

show abstract

Section: Introductionmentioning

confidence: 99%

Iterative Learning Control Integrated with Model Predictive Control for Real-Time Disturbance Rejection of Batch Processes

Lee

2017

J. Chem. Eng. Japan / JCEJ

View full text Add to dashboard Cite

show abstract

“…It can be considered as a two-dimensional model that is a function of the time evolution of the process behavior during the considered operation (index ) and the operation number (index ). This partial knowledge of the process behavior is essentially based on energy balance equations of the reacting mixture and of the reactor jacket wall [20]. These energy balance equations will be considered in order to derive the process outputs during the whole batch operation [25].…”

Section: Modelingmentioning

confidence: 99%

“…Hence, from the least-square formula, the expression of the optimal sequence ( ) is readily obtained (19) and the first component of the vector is considered as the instantaneous correction term to the control value computed at the previous batch. The control value is given by (20) …”

Section: B Control Synthesismentioning

confidence: 99%

See 1 more Smart Citation

Application of iterative learning control to an exothermic semibatch chemical reactor

Mezghani¹,

Roux

Cabassud

et al. 2002

IEEE Trans. Contr. Syst. Technol.

View full text Add to dashboard Cite

This work focuses on the temperature control of a semibatch chemical reactor used for fine chemicals production. Such a reactor is equipped with a heating/cooling system composed of different thermal fluids. Without extensive modeling investigations, a feedback-feedforward control strategy is proposed for ensuring the tracking performance of the desired temperature profile. Such a strategy is derived from a family of the iterative learning control (ILC) algorithms named batch model predictive control (BMPC). Learning is achieved without requiring a detailed knowledge of the system, which may be affected by unknown but repetitive disturbances. The learning control solution is based on the minimization of a linear quadratic cost function. The synthesis of the proposed strategy is studied, and improvements of the algorithm features are proposed. First, guaranteed convergence of the algorithm is illustrated in a few experimental runs. Second, some practical considerations for the removal of high-frequency disturbance effects are outlined to improve the achieved performance. Third, a robust supervisory control procedure is employed to choose the right fluid and to reduce the superfluous fluid changeovers, mainly when different fluids are available. Finally, experimental results are presented to illustrate the practical appeal and effectiveness of the proposed scheme.

show abstract

“…[1][2][3][4] The optimal operation of a batch process aims to produce an end product that meets quality specifications. [5][6][7] Since the batch process is repetitive in nature, it is often designed to improve product quality by learning from operation to operation, [1][2][3][4][5][6][7] which is the essence of iterative learning control (ILC). [8] Hence, some literature [1][2][3][4][5][6][9][10][11][12] has explored the application of ILC into batch manufacturing processes.…”

mentioning

confidence: 99%

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

et al. 2017

View full text Add to dashboard Cite

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 datadriven 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.

show abstract

Robust iterative learning control of an exothermic semi-batch chemical reactor

Cited by 22 publications

References 14 publications

Iterative Learning Control Integrated with Model Predictive Control for Real-Time Disturbance Rejection of Batch Processes

Iterative Learning Control Integrated with Model Predictive Control for Real-Time Disturbance Rejection of Batch Processes

Application of iterative learning control to an exothermic semibatch chemical reactor

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

Contact Info

Product

Resources

About