Neural Network Model Predictive Control Applied to a Fed-batch Sugar Crystallization (original) (raw)
Related papers
Neural Network-Based Control Strategies Applied to a Fed-Batch Crystallization Process
2007
This paper is focused on issues of process modeling and two model based control strategies of a fed-batch sugar crystallization process applying the concept of artificial neural networks (ANNs). The control objective is to force the operation into following optimal supersaturation trajectory. It is achieved by manipulating the feed flow rate of sugar liquor/syrup, considered as the control input. The
2007
This paper is focused on issues of nonlinear dynamic process modeling and model-based predictive control of a fed-batch sugar crystallization process applying the concept of artificial neural networks as computational tools. The control objective is to force the operation into following optimal supersaturation trajectory. It is achieved by manipulating the feed flow rate of sugar liquor/syrup, considered as the control input. A feed forward neural network (FFNN) model of the process is first built as part of the controller structure to predict the process response over a specified (prediction) horizon. The predictions are supplied to an optimization procedure to determine the values of the control action over a specified (control) horizon that minimizes a predefined performance index. The control task is rather challenging due to the strong nonlinearity of the process dynamics and variations in the crystallization kinetics. However, the simulation results demonstrated smooth behavio...
Nonlinear predictive control based on artificial neural network model for industrial crystallization
Journal of Food Engineering, 2010
This paper illustrates the benefits of a nonlinear model based predictive control (NMPC) strategy for setpoint tracking control of an industrial crystallization process. A neural networks model is used as internal model to predict process outputs. An optimization problem is solved to compute future control actions taking into account real-time control objectives. Furthermore, a more suitable output variable is used for process control: the mass of crystals in the solution is used instead of the traditional electrical conductivity. The performance of the NMPC implementation is assessed via simulation results based on industrial data.
Nonlinear model predictive control strategies applied to a fed-bath sugar crystallizer
The present work is focused on a comparative study of two nonlinear MPC (NMPC) control schemes implemented to a fed-batch sugar crystallization process -i) NMPC that does not exploit the batch nature of the process (termed as classical NMPC) and ii) the batch NMPC that takes into account the end-point control objectives. They are also compared with the classical PI controller and a linear MPC scheme. Two main scenarios are considered: a nominal case without disturbances and a case with disturbances and variations in the initial conditions. The results demonstrate that the batch NMPC outperforms the other control structures but to the expense of high computational cost.
Linear Model Predictive Control Strategies Applied to a Batch Sugar Crystallizer
2006
The study of four structures of linear model predictive control (LMPC) for a batch white sugar crystallization process is reported in this paper. Two SISO and two MIMO control schemes were compared with respect to the final process quality achieved. The linear models required in the controller structures were extracted applying two identification alternatives. The SISO cases seem to guarantee more satisfactory end point quality of the process. However, only the LMPC of the supersaturation manipulating the steam flowrate makes feasible all conflicting control objectives.
Nonlinear MPC for fed-batch multiple stages sugar crystallization
Chemical Engineering Research and Design, 2011
This paper addresses the issue of developing feasible advanced control strategies for the operation of industrial fedbatch multi-stage sugar crystallization processes. The operation of such processes poses very challenging problems mainly those inherent to its batch nature and also those due to the difficulties in measuring key process variables. Inadequate control policies lead to out-of-spec batches, with consequent losses resulting from the need of product recycling. In order to address these problems, a modification of the general Nonlinear Model Predictive Control (NMPC) is proposed in this paper, where the NMPC is executed only when the tracking error is outside a pre-specified bound. Once the error converges towards the˛-strip, the NMPC is switched off and the control action is kept constant. In order to further reduce the complexity of the control system, the proposed modification, termed Error Tolerant MPC (ETMPC), is provided with a Recurrent Neural Network (RNN) predictive model. The ETMPC + RNN control scheme was extensively tested on a crystallizer dynamic simulator, tuned with data from two industrial units, and compared with the classical NMPC and PI strategy. The results demonstrate that both NMPC and ETMPC controllers lead to improved end point process specifications, when compared with the PI controller. The explicit introduction of the error tolerance in the optimization relaxes the computational burden and can complement several other suggestions in the literature for feasible industrial real time control.
Intelligent Predictive Control - Application to Scheduled Crystallization Processes
2009 International Conference on Adaptive and Intelligent Systems, 2009
The purpose of this paper is twofold. On the one hand, we propose a modification of the general Model Predictive Control (MPC) approach where a prespecified tracking error is tolerated. The introduction of error tolerance (ET) in the MPC optimization algorithm reduces considerably the average duration of each optimization step and makes the MPC computationally more efficient and attractive for industrial applications. On the other hand a challenging scheduled crystallization process serves as a case study to show the practical relevance of the new intelligent predictive control. Comparative tests with different control policies are performed: i) Classical MPC with analytical or Artificial Neural Network (ANN) process model; ii) ET MPC with analytical or ANN process model; iii) Proportional-Integral (PI) control. Besides the computational benefits of ET MPC, the integration of ANN into the ET MPC brings substantial improvements of the final process performance measures and further relaxes the computational demands.
Neural networks for model predictive control
This paper is focused on developing a model predictive control (MPC) based on recurrent neural network (NN) models. Two regression NN models suitable for prediction purposes are proposed. In order to reduce their computational complexity and to improve their prediction ability, issues related with optimal NN structure (lag space selection, number of hidden nodes), pruning techniques and identification strategies are discussed. The NN-based MPC and the traditional PI (Proportional-Integral) control are tested in the presence of process disturbances on a crystallizer dynamic simulator.
Hybrid model predictive control of a sugar end section
2006
This paper deals with the MPC control of an industrial hybrid process where continuous and batch units operate jointly: the crystallization section of a sugar factory. The paper describes a plant-wide predictive controller that takes into account, both, the continuous objectives and manipulated variables, as well as the ones related to the scheduling of the batch units. The MPC is formulated avoiding the use of integer variables, so that a NLP optimization technique could be applied. Simulation results of the controller operation are provided * .