Neural Network Predictive Control Research Articles

NN-adaptive predictive control for a class of discrete-time nonlinear systems with input-delay

An adaptive neural network (NN) predictive control algorithm is proposed for a class of uncertain discrete-time nonlinear systems with input-delay. For approximate the nonlinear part of the uncertain discrete-time nonlinear system, the recurrent neural network (RNN) is employed. A version of discrete-time NN weight update rule is given. For compensate the reconstructive error of the RNN and the external disturbances, the robust control term is designed. In order to compensate the time-delay in input channel, a predictive control algorithm is proposed. Then the synchronous tracking objective can be realized. An illustrative example is provided to demonstrate the effectiveness of the proposed control strategy.

HYBRID INTELLIGENT CONTROLLERS FOR A MULTIPLE DRUG DELIVERY SYSTEM IN ACUTE HEART FAILURE

Regulating the dynamic responses to multiple therapeutic agents in cases of heart failure is difficult owing to time-variant changes in drug sensitivity and interaction. To address this problem, a multiple controller based on adaptive neural network (NN) predictive control has been developed for unexpected drug responses related to cardiac output and arterial pressure. However, the control speed may be slower than that in traditional controllers because of the real-time learning process for the NN. Moreover, a proportional-integral-derivative (PID) controller alone cannot automatically update the PID parameters during drug administration. This study, therefore, aimed to make hybrid intelligent (fuzzy or NN-based PID) controllers and to evaluate the control performance during multiple drug therapy in unexpected physiological responses of heart failure. The hybrid intelligent controllers were compared with the previous PID or NN controller, and they realized robust and quick control regardless of unexpected responses and acute disruptions.

DEPTH CONTROL OF AN UNDERWATER REMOTELY OPERATED VEHICLE USING NEURAL NETWORK PREDICTIVE CONTROL

This paper investigates the depth control of an unmanned underwater remotely operated vehicle (ROV) using neural network predictive control (NNPC). The NNPC is applied to control the depth of the ROV to improve the performances of system response in terms of overshoot. To assess the viability of the method, the system was simulated using MATLAB/Simulink by neural network predictive control toolbox. In this paper also investigates the number of data samples (1000, 5000 and 10,000) to train neural network. The simulation reveals that the NNPC has the better performance in terms of its response, but the execution time will be increased. The comparison between other controller such as conventional PI controller, Linear Quadratic Regulation (LQR) and fuzzy logic controller also covered in this paper where the main advantage of NNPC is the fastest system response on depth control.

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.

Model Identification of an Underwater Remotely Operated Vehicle Using System Identification Approach Based on NNPC

This paper described the development of modeling of an unmanned underwater vehicle (UUV) using system identification toolbox based on neural network model. The set of data based on neural network model generated by open-loop model of UUV and the input-output data produced using neural network predictive control technique. The model of UUV is an underwater Remotely Operated Vehicle (ROV) will be used in this study. Open-loop model of ROV created using system identification technique with implemented in real time experiment for open-loop system. Two data will be used such as the input and output neural network data for validation and training for infer a model of the ROV using system identification toolbox. The data re-generated using graph digitizer software. The accuracy of this software almost 90%. Then, the model obtained in this system will be controlled using conventional PID controller in MATLAB Simulink. The comparison between two models from different techniques of the ROV will be described. When the number of samples used in this project reduced, the best fit will be increased. A model obtained based on neural network model is acceptable to use in simulation and will be improved the best fit when reduced number of samples.

Control of a heat exchanger using neural network predictive controller combined with auxiliary fuzzy controller

The paper presents an advanced control strategy that uses the neural network predictive controller and the fuzzy controller in the complex control structure with an auxiliary manipulated variable. The controlled tubular heat exchanger is used for pre-heating of petroleum by hot water. The heat exchanger is modelled as a nonlinear system with the interval parametric uncertainty. The set point tracking and the disturbance rejection using intelligent control strategies are investigated. The control objective is to keep the outlet temperature of the pre-heated petroleum at a reference value. Simulations of control of the tubular heat exchanger are done in the Matlab/Simulink environment. The complex control structure with two controllers is compared with the conventional PID control, fuzzy control and NNPC. Simulation results confirm the effectiveness and superiority of the complex control structure combining the NNPC with the auxiliary fuzzy controller.

Hot Blast Stove Temperature Control System Based on Neural Network Predictive Control

Chain type coal-fired hot blast furnace boiler has a strong coupling, large delay, large inertia characteristics. Control effect of control method of mathematic modeling method and the classical routine of it is very difficult to produce the ideal. The predictive control theory combined with neural network theory. Through the model correction and rolling optimization control method of the system is good to overcome the effects of model error and time-varying process. The experimental results showed that neural network predictive control system is improved effectively the static precision and dynamic characteristic. It has better practicability of boiler temperature of this kind of large time delay system.

Control of a Heat Exchanger Using Neural Network Predictive Controller and Auxiliary Fuzzy Controller

Control of a Heat Exchanger Using Neural Network Predictive Controller and Auxiliary Fuzzy Controller

Neural Network Predictive Controller Based Nonlinearity Identification Case Study: Nonlinear Process Reactor - CSTR

In the last decades, a substantial amount of research has been carried out on identification of nonlinear processes. Dynamical systems can be better represented by nonlinear models, which illustrate the global behavior of the nonlinear process reactor over the entire range. CSTR is highly nonlinear chemical reactor. A compact and resourceful model which approximates both linear and nonlinear component of the process is of highly demand. Process modeling is an essential constituent in the growth of sophisticated model-based process control systems. Driven by the contemporary economical needs, developments in process design point out that deliberate operation requires better models. The neural network predictive controller is very efficient to identify complex nonlinear systems with no complete model information. Closed loop method is preferred because it is sensitive to disturbances, no need identify the transfer function model of an unstable system. In this paper identification nonlinearities for a nonlinear process reactor CSTR is approached using neural network predictive controller. KEYWORDS Continuous Stirred Tank Reactor, Multi Input Multi Output, Neural Networks, Chebyshev Neural Networks, Predictive Controller.

ChemInform Abstract: Thermal Stress Management of a Solid Oxide Fuel Cell Using Neural Network Predictive Control

AbstractReview: 53 refs.

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

Neural network and support vector machine predictive control of tert-amyl methyl ether reactive distillation column

An algorithm of model predictive control based on artificial neural network and least-square support vector machine method is presented for a class of industrial process with strong nonlinearity such as tert-amyl methyl ether (TAME). Integral constant is added to improve the performance of the controller. In the present work, two different control methodologies neural network predictive control (NNPC) and support vector machine-based predictive control (SVMPC) are implemented and compared with a conventional proportional-integral-derivative (PID) control methodology to a TAME reactive distillation column. The simulation result shows that both NNPC and SVMPC gives better control performance than PID for set-point change as well as for load change of±10% in methanol feed flow rate and molar ratio of methanol to isoamylene in reactor effluent feed.

Predictive Control Strategy Based on Extreme Learning Machine for Path-Tracking of Autonomous Mobile Robot

In this paper, we propose a novel nonlinear predictive control strategy based on an extreme learning machine to address the path-tracking control problem of wheeled mobile robots in the presence external disturbances. The hybrid chaotic optimization algorithm (HCOA), which can avoid being trapped in local minima and improve convergence in dealing with the large space and high-dimension optimization problems, is used to perform real-time nonlinear minimization of the cost function of a mobile robot to enhance the control accuracy. The proposed improved bidirectional extreme learning machine is employed to model the mobile robot plant and estimate future plant output. The experimental results of tracking the automation mobile robot indicate that the proposed controller can provide more accuracy and faster tracking performance than traditional neural network predictive controllers.

A hybrid-driven underwater glider model, hydrodynamics estimation, and an analysis of the motion control

An autonomous hybrid-driven glider is a new class of autonomous underwater glider that integrates the concept of a buoyancy-driven underwater glider and a conventional autonomous underwater vehicle (AUV). This glider has multi-functionality that enables it to overcome the speed and maneuverability limitations of buoyancy-driven gliders. Thus, this paper presents a mathematical model of the hybrid-driven glider, an estimation of the hydrodynamics and an analysis of the motion control. A Newtonian approach has been used to model the glider under the influence of water current. In addition, the hydrodynamics were estimated by using a Strip theory and a computational fluid dynamic (CFD). In order to analyze the motion of the glider, a Neural Network Predictive Control (NNPC) has been designed and its performance has been compared with the Model Predictive Control (MPC) and Linear Quadratic Regulator (LQR). The simulation results demonstrated that the Neural Network Predictive Control (NNPC) produced better control performance than the performance of the MPC and LQR when dealing with disturbances. In addition, the results also demonstrate the hydrodynamic response of the glider over the velocity, angle of attack and sideslip angle.

Hardware Implementation of the Neural Network Predictive Controller for Coupled Tank System

In this paper, a neural network based predictive controller is designed for controlling the liquid level of the coupled tank system. The controlled process is a nonlinear system; therefore, a nonlinear prediction method can be a better match in a predictive control strategy. The neural network predictive controller that is discussed in this paper uses a neural network model of a nonlinear plant to predict future plant performance. The simulation results are compared with PID control. The results show that the effectiveness of using the neural predictive controller for the coupled tank system. The Simulink Toolbox in MATLAB has been used to simulate the controlled system with the proposed controller. The VHDL has been used to describe the implementation of neural controller. Xilinx ISE Project Navigator Version 10.1 is used to obtain the compilation and timing test results as well as the synthesized design. The hardware implementation of the neural network predictive controller using FPGA board is proposed. To make sure that the FPGA board works like the simulated neural predictive controller, MATLAB programme is used to compare between the set of the data that are obtained from the ModelSim program and the set of the data that are obtained from the MATLAB Simulink model. Simulation results show that the FPGA board can be used as neural predictive controller for controlling the liquid level of the coupled tank system.

Nonlinear Recurrent Neural Network Predictive Control for Energy Distribution of a Fuel Cell Powered Robot

This paper presents a neural network predictive control strategy to optimize power distribution for a fuel cell/ultracapacitor hybrid power system of a robot. We model the nonlinear power system by employing time variant auto-regressive moving average with exogenous (ARMAX), and using recurrent neural network to represent the complicated coefficients of the ARMAX model. Because the dynamic of the system is viewed as operating- state- dependent time varying local linear behavior in this frame, a linear constrained model predictive control algorithm is developed to optimize the power splitting between the fuel cell and ultracapacitor. The proposed algorithm significantly simplifies implementation of the controller and can handle multiple constraints, such as limiting substantial fluctuation of fuel cell current. Experiment and simulation results demonstrate that the control strategy can optimally split power between the fuel cell and ultracapacitor, limit the change rate of the fuel cell current, and so as to extend the lifetime of the fuel cell.

A New Methodology for Simulation and Optimization of CO2 Sequestration in a Saline Aquifer Using Artificial Neural Network and Model Predictive Control Approach

Carbon dioxide sequestration efforts are the subject of intense research due to growing international concerns over CO2 emissions critically affecting the global climate change. Extensive studies are being done to determine the extent of the CO2 injection so that CO2 storage is secure and environmentally acceptable. In this work, CO2 injection into a saline aquifer is used to simulate an aquifer in Eclipse-100 using a black oil reservoir simulator whose consequent behavior is predicted by an artificial neural network, being trained via the Levenberg-Marquardt algorithm. A generalized predictive control scheme is accommodated in the proposed design methodology to determine the optimal injection rates during the CO2 injection process, leading to an ultimate goal of maximized residual trapping or equivalently minimized structural trapping of carbon dioxide. This will eventually maximize the total volume of the injected CO2, while diminishing the risk of its leakage to the atmosphere due to a fail in the integrity of the formation cap rock. A set of diverse test scenarios has been organized to demonstrate the capabilities of the proposed methodology to tackle the probable practical considerations.

Brief Study of Identification System Using Regression Neural Network Based on Delay Tap Structures

Abstractthis investigation aims to provide an overview and investigations of non-linear identification system using neural network approach. Nowadays, a lot of neural network approach was done to provide a satisfying identification system. Backpropagation scheme as the mainstream approach for developing identification system has several limitations such as training data, computation time, architecture, optimization technique for weight value update, and many others. Regression Neural Network which is found by specht on 1990 contains more advantages compare with backpropagation scheme. With the improvement of computation time, architecture, and robustness of this model and provided 90% of effectiveness, promising a good prospective to develop non-linear identification system. For future works, it can be implemented in neural network predictive model control system and another control scheme based on identification system approach.

Thermal stress management of a solid oxide fuel cell using neural network predictive control

In SOFC (solid oxide fuel cell) systems operating at high temperatures, temperature fluctuation induces a thermal stress in the electrodes and electrolyte ceramics; therefore, the cell temperature distribution is recommended to be kept as constant as possible. In the present work, a mathematical model based on first principles is presented to avert such temperature fluctuations. The fuel cell running on ammonia is divided into five subsystems and factors such as mass/energy/momentum transfer, diffusion through porous media, electrochemical reactions, and polarization losses inside the subsystems are presented. Dynamic cell-tube temperature responses of the cell to step changes in conditions of the feed streams is investigated. The results of simulation indicate that the transient response of the SOFC is mainly influenced by the temperature dynamics. It is also shown that the inlet stream temperatures are associated with the highest long term start-up time (467 s) among other parameters in terms of step changes. In contrast the step change in fuel velocity has the lowest influence on the start-up time (about 190 s from initial steady state to the new steady state) among other parameters. A NNPC (neural network predictive controller) is then implemented for thermal stress management by controlling the cell tube temperature to avoid performance degradation by manipulating the temperature of the inlet air stream. The regulatory performance of the NNPC is compared with a PI (proportional–integral) controller. The performance of the control system confirms that NNPC is a non-linear-model-based strategy which can assure less oscillating control responses with shorter settling times in comparison to the PI controller.

The Predictive Control Method of VAV Air Conditioning System

Aiming at the characteristics which variable air volume air conditioning system is multi-variable, nonlinear and uncertain system, normal fuzzy neural network is hard to meet the requirements which dynamic control of multi-variable. In this paper, we put forward a recursive neural network predictive control strategy based on wavelet neural network model. Through recursive wavelet neural network predictor on line established controlled object’s mathematical model, and using Elman neural network controller on line corrected information we get, thus to improve control effect. The simulation results show that recursive wavelet neural network predictive control has stronger robustness and adaptive ability, high control precision, better and reliable control effect and other advantages. DOI: http://dx.doi.org/10.11591/telkomnika.v11i9.3244