Neural Network Model Predictive Control Research Articles

A Combined Adaptive Neural Network and Nonlinear Model Predictive Control for Multirate Networked Industrial Process Control

This paper investigates the multirate networked industrial process control problem in double-layer architecture. First, the output tracking problem for sampled-data nonlinear plant at device layer with sampling period T(d) is investigated using adaptive neural network (NN) control, and it is shown that the outputs of subsystems at device layer can track the decomposed setpoints. Then, the outputs and inputs of the device layer subsystems are sampled with sampling period T(u) at operation layer to form the index prediction, which is used to predict the overall performance index at lower frequency. Radial basis function NN is utilized as the prediction function due to its approximation ability. Then, considering the dynamics of the overall closed-loop system, nonlinear model predictive control method is proposed to guarantee the system stability and compensate the network-induced delays and packet dropouts. Finally, a continuous stirred tank reactor system is given in the simulation part to demonstrate the effectiveness of the proposed method.

Modified control algorithms for patients with type 1 diabetes mellitus undergoing exercise

In this article, two previously published control algorithms for type 1 diabetes mellitus (T1DM) are modified for patients undergoing exercise. These two control algorithms include the hybrid neural network model predictive control employed on continuous subcutaneous insulin infusion (CSII) therapy, and the fuzzy-logic-based supervisor applied to multiple daily injections (MDI) therapy. A simulation model incorporating a well-acknowledged meal glucose–insulin model and an additional model of physical activity is employed to create virtual subjects for testing. The main notion of the modifications of the control algorithms is to subtract extra carbohydrate for aerobic exercise (ExCarbs) to prevent the exercise-induced hypoglycemia. This residual of CHO, after subtracting ExCarbs from an estimated carbohydrate (CHO) of a meal, is a more appropriate quantity of the CHO which should be compensated by external insulin. The two modified control algorithms are tested for subjects with various intensities and durations of exercise. Simulation results of the virtual subjects show that the algorithms are effective to prevent exercise-induced hypoglycemia. In addition, the simulation results also reveal that the modified methodologies are still robust with blood glucose (BG) level of subjects maintaining within safe region.

Algorithm of Multi Sensor Data Fusion based on BP Neural Network and Multi-scale Model Predictive Control

Algorithm of Multi Sensor Data Fusion based on BP Neural Network and Multi-scale Model Predictive Control

Survey on nonlinear reconfigurable flight control

An overview on nonlinear reconfigurable flight control approaches that have been demonstrated in flight-test or highfidelity simulation is presented.Various approaches for reconfigurable flight control systems are considered,including nonlinear dynamic inversion,parameter identification and neural network technologies,backstepping and model predictive control approaches.The recent research work,flight tests,and potential strength and weakness of each approach are discussed objectively in order to give readers and researchers some reference.Finally,possible future directions and open problems in this area are addressed.

Constant Cutting-Power Control of Shearer Based on Neural Network Model Predictive Control

To solve the problem of constant power control of shearer cutting machine, the nonlinear predictive control method based on Neural Network was proposed in this thesis. In the method, the cutting current was used to identify the cutting load, and the Neural Network was used to predict and control the traction speed. A Neural Network model was built by the current and speed to control the cutting power of shearer. In MATLAB, the field data was used to simulate and the simulation verify the proposed scheme is better than PID method.

On the Structural Optimization of a Neural Network Model Predictive Controller

This paper provides a new algorithm for tuning the two most effective parameters in nonlinear model predictive control (NMPC). Tuning is performed in two steps. First, a new method based on Barron’s formula, the bicoherence nonlinearity test, and inphase-quadrature demodulation is proposed to determine the number of hidden layer neurons in a two-layer neural network. In the second step, a fuzzy algorithm is introduced to tune the input weight matrix in the objective function to make the tuning problem more practical and precise. To show the effectiveness of the proposed method, several examples are discussed including a simple flow process and a more complex pH neutralization problem. The method is also evaluated in the laboratory scale pressure and level processes. It is shown that the proposed method leads to tuning the number of neurons and the weight matrix with an acceptable performance.

Predictive Control and Simulation for Variable-Pitch Wind Turbines

In order to avoid the limitations of traditional mechanism modeling method, a neural network (NN) model of variable - pitch wind turbine is built by the NN modeling method based on field data. Then considering that from wind turbine’s startup to grid integration, the generator speed must be controlled to rise to the synchronous speed smoothly and precisely, a neural network model predictive control (NNMPC) strategy based on the small-world optimization algorithm (SWOA) is proposed. Simulation results show that the strategy can forecast the change of generator rotational speed based on the wind speed disturbance, making the controller act ahead to eliminate the impact of system delay. Furthermore, the system output can track the reference trajectory well, making sure that the system can connect the electricity grid steadily.

A Hybrid Neural Network Model Predictive Control with Zone Penalty Weights for Type 1 Diabetes Mellitus

In this paper, a hybrid neural network model is developed to predict and control the blood glucose (BG) of the patient who has type 1 diabetes mellitus (T1DM). The proposed model consists of two parts: a linear finite impulse response (FIR) model and a nonlinear autoregressive exogenous input (NARX) network. A recently developed and well-acknowledged meal simulation model of the glucose-insulin system for T1DM is employed to create virtual subjects. Data from virtual subjects are used to identify an intermediate physiological model, and then our proposed hybrid model is trained and validated based on this intermediate model. The key features of the resulting hybrid model are that it reveals satisfactory accuracy of long-term prediction and does not require an immeasurable state for model initialization. The developed hybrid model is then embedded in a nonlinear model predictive control (MPC) controller with zone penalty weights, and this closed-loop controller is implemented on these virtual subjects for simulation-based preclinical testing. The results show that promising glycemic control performance can be achieved. Moreover, this overall BG control methodology is easily portable and has the ability to arbitrarily start the therapeutic control at any initial point.

Model predictive control of dissolved oxygen concentration based on a self-organizing RBF neural network

The dissolved oxygen (DO) concentration in activated sludge wastewater treatment processes (WWTPs) is difficult to control because of the complex nonlinear behavior involved. In this paper, a self-organizing radial basis function (RBF) neural network model predictive control (SORBF-MPC) method is proposed for controlling the DO concentration in a WWTP. The proposed SORBF can vary its structure dynamically to maintain prediction accuracy. The hidden nodes in the RBF neural network can be added or removed on-line based on node activity and mutual information (MI) to achieve the appropriate network complexity and the necessary dynamism. Moreover, the convergence of the SORBF is analyzed in both the dynamic process phase and the phase following the modification of the structure. Finally, the SORBF-MPC is applied to the Benchmark Simulation Model 1 (BSM1) WWTP to maintain the DO concentration. The results show that SORBF-MPC effectively provides process control. The performance comparison also indicates that the proposed model's predictive control strategy yields the most accurate for DO concentration, better effluent qualities, and lower average aeration energy (AE) consumption.

Neural network model predictive control of a styrene polymerization plant: online testing using an electronic worksheet

AbstractThe batch styrene polymerization process presents transient and nonlinear temperature behavior. In this work, manual control and open loop experiments were carried out in order to build a process knowledge database. Initially, a cascade feedback control loop was implemented by manipulating the thyristor unit of the electrical heater in the thermal fluid tank. Aiming at the MPC development, algebraic equations of a neural network and its adjusted parameters were implemented in an electronic worksheet. Every five seconds, the worksheet was updated with measurements (reactor temperature, thermal fluid temperature and thyristor position) by means of the OLE for the Process Control protocol (OPC). The one-step-ahead temperature prediction was then employed in the objective function of the worksheet solver which used Visual Basic Applications programming. The manipulated variable action was then calculated and sent to the process. A hybrid controller (cascade feedback and MPC) outperformed the pure strategies since the time-integral performance indexes, IAE and ITAE, were reduced by around 22 % and 32 %, respectively. Methodology for the Model Predictive Control presented in this study was considered feasible because the solver of Microsoft Office Excel (2007) is very friendly, easy to understand and ready to implement using VBA.

NEURAL NETWORK PREDICTIVE CONTROL BASED POWER SYSTEM STABILIZER

The present study investigates the power system stabilizer based on neural predictive control for improving power system dynamic performance over a wide range of operating conditions. In this study a design and application of the Neural Network Model Predictive Controller (NN-MPC) on a simple power system composed of a synchronous generator connected to an infinite bus through a transmission line is proposed. The synchronous machine is represented in detail, taking into account the effect of the machine saliency and the damper winding. Neural network model predictive control combines reliable prediction of neural network model with excellent performance of model predictive control using nonlinear Levenberg-Marquardt optimization. This control system is used the rotor speed deviation as a feedback signal. Furthermore, the using performance system of the proposed controller is compared with the system performance using conventional one (PID controller) through simulation studies. Digital simulation has been carried out in order to validate the effectiveness proposed NN-MPC power system stabilizer for achieving excellent performance. The results demonstrate that the effectiveness and superiority of the proposed controller in terms of fast response and small settling time.

Model Predictive Inferential Control of a Distillation Column

Typical production objectives in distillation process require the delivery of products whose compositions meet certain specifications. The distillation control system, therefore, must hold product compositions as near the set points as possible in faces of upset. In this project, inferential model predictive control, that utilizes an artificial neural network estimator and model predictive controller, is developed for an industrial multicomponent distillation column. First, composition control by direct measuring composition is used. This method because of large sampling delay has a poor performance. The selection of the temperature measurement points is done for indirect control of the column. The use of temperature loop leads to an offset in the composition; due to the fact that the temperature set-point must be changed when feed disturbances occurred. An artificial neural network estimator is designed to estimate the product compositions from tray temperature measurements. A model predictive controller is used to control column composition based on composition estimates. The performance of the developed inferential model predictive control system is tested for set-point tracking and load rejection.

Control of polystyrene batch reactors using neural network based model predictive control (NNMPC): An experimental investigation

Controlling batch polymerization reactors imposes great operational difficulties due to the complex reaction kinetics, inherent process nonlinearities and the continuous demand for running these reactors at varying operating conditions needed to produce different polymer grades. Model predictive control (MPC) has become the leading technology of advanced nonlinear control adopted for such chemical process industries. The usual practice for operating polymerization reactors is to optimize the reactor temperature profile since the end use properties of the product polymer depend highly on temperature. This is because the end use properties of the product polymer depend highly on temperature. The reactor is then run to track the optimized temperature set-point profile. In this work, a neural network-model predictive control (NN-MPC) algorithm was implemented to control the temperature of a polystyrene (PS) batch reactors and the controller set-point tracking and load rejection performance was investigated. In this approach, a neural network model is trained to predict the future process response over the specified horizon. The predictions are passed to a numerical optimization routine which attempts to minimize a specified cost function to calculate a suitable control signal at each sample instant. The performance results of the NN-MPC were compared with a conventional PID controller. Based on the experimental results, it is concluded that the NN-MPC performance is superior to the conventional PID controller especially during process startup. The NN-MPC resulted in smoother controller moves and less variability.

Neural predictive controller of a two-area load frequency control for interconnected power system

The present paper investigates the load-frequency control (LFC) for improving power system dynamic performance over a wide range of operating conditions. This study proposed design and application of the neural network model predictive controller (NN-MPC) on two-area load frequency power systems. Neural network model predictive control (NN-MPC) combines reliable prediction of neural network with excellent performance of model predictive control using nonlinear Levenberg–Marquardt optimization. The controller used the local power area error deviation as a feedback signal. To validate the effectiveness of the proposed controller, two-area power system is simulated over a wide range of operating conditions and system parameters change. Further, the performance of the proposed controller is compared with a fuzzy logic controller (FLC) through simulation studies. Obtained results demonstrate the effectiveness and superiority of the proposed approach.

Elevating Model Predictive Control Using Feedforward Artificial Neural Networks: A Review

Process control in the field of chemical engineering has always been a challenging task for the chemical engineers. Hence, the majority of processes found in the chemical industries are non-linear and in these cases the performance of the linear models can be inadequate. Recently a promising alternative modelling technique, artificial neural networks (ANNs), has found numerous applications in representing non-linear functional relationships between variables. A feedforward multi-layered neural network is a highly connected set of elementary non-linear neurons. Model-based control techniques were developed to obtain tighter control. Many model-based control schemes have been proposed to incorporate a process model into a control system. Among them, model predictive control (MPC) is the most common scheme. MPC is a general and mathematically feasible scheme to integrate our knowledge about a target, process controller design and operation, which allows flexible and efficient exploitation of our understanding of a target, and thus produces optimal performance of a system under various constraints. The need to handle some difficult control problems has led us to use ANN in MPC and has recently attracted a great deal of attention. The efficacy of the neural predictive control with the ability to perform comparably to the non linear neural network strategy in both set point tracking and disturbance rejection proves to have less computation expense for the neural predictive control. The neural network model predictive control (NNMPC) method has less perturbations and oscillations when dealing with noise as compared to the PI controllers.

Neural network based model predictive control for a steel pickling process

A multi-layer feedforward neural network model based predictive control scheme is developed for a multivariable nonlinear steel pickling process in this paper. In the acid baths three variables under controlled are the hydrochloric acid concentrations. The baths exhibit the normal features of an industrial system such as nonlinear dynamics and multi-effects among variables. In the modeling, multiple input, single-output recurrent neural network subsystem models are developed using input–output data sets obtaining from mathematical model simulation. The Levenberg–Marquardt algorithm is used to train the process models. In the control (MPC) algorithm, the feedforward neural network models are used to predict the state variables over a prediction horizon within the model predictive control algorithm for searching the optimal control actions via sequential quadratic programming. The proposed algorithm is tested for control of a steel pickling process in several cases in simulation such as for set point tracking, disturbance, model mismatch and presence of noise. The results for the neural network model predictive control (NNMPC) overall show better performance in the control of the system over the conventional PI controller in all cases.

The design and optimization for light-algae bioreactor controller based on Artificial Neural Network-Model Predictive Control

As a subsystem of the bioregenerative life support system (BLSS), light-algae bioreactor (LABR) has properties of high reaction rate, efficiently synthesizing microalgal biomass, absorbing CO 2 and releasing O 2 , so it is significant for BLSS to provide food and maintain gas balance. In order to manipulate the LABR properly, it has been designed as a closed-loop control system, and technology of Artificial Neural Network–Model Predictive Control (ANN-MPC) is applied to design the controller for LABR in which green microalgae, Spirulina platensis is cultivated continuously. The conclusion is drawn by computer simulation that ANN-MPC controller can intelligently learn the complicated dynamic performances of LABR, and automatically, robustly and self-adaptively regulate the light intensity illuminating on the LABR, hence make the growth of microalgae in the LABR be changed in line with the references, meanwhile provide appropriate damping to improve markedly the transient response performance of LABR.

Disturbance Rejection in Neural Network Model Predictive Control

Neural Network Model Predictive Control (NN-MPC) combines reliable prediction of neural network with excellent performance of model predictive control using nonlinear Levenberg-Marquardt optimization. It is shown that this structure is prone to steady-state error when external disturbances enter or actual system varies from its model. In this paper, these model uncertainties are taken into account using a disturbance model with iterative learning which adaptively change the learning rate to treat gradual effect of the model mismatch differently from the drastic changes of external disturbance. Then, a high-pass filter on error signal is designed to distinguish disturbances from model mismatches. Practical implementation results as well as simulation results demonstrate good performance of the proposed control method.

A neural network model predictive controller

A neural network controller is applied to the optimal model predictive control of constrained nonlinear systems. The control law is represented by a neural network function approximator, which is trained to minimize a control-relevant cost function. The proposed procedure can be applied to construct controllers with arbitrary structures, such as optimal reduced-order controllers and decentralized controllers.

Neural-Networks-Based Feedback Linearization versus Model Predictive Control of Continuous Alcoholic Fermentation Process

In this work advanced nonlinear neural networks based control system design algorithms are adopted to control a mechanistic model for an ethanol fermentation process. The process model equations for such systems are highly nonlinear. A neural network strategy has been implemented in this work for capturing the dynamics of the mechanistic model for the fermentation process. The neural network achieved has been validated against the mechanistic model. Two neural network based nonlinear control strategies have also been adopted using the model identified. The performance of the feedback linearization technique was compared to neural network model predictive control in terms of stability and set point tracking capabilities. Under servo conditions, the feedback linearization algorithm gave comparable tracking and stability. The feedback linearization controller achieved the control target faster than the model predictive one but with vigorous and sudden controller moves.