Self-tuning Predictive Control Research Articles

Self‐tuning predictive control applicable to ship magnetic levitation damping device

In the ship design, there are strict vibration-proof requirements for precision instruments. Therefore, a ship repulsive magnetic levitation damping device is designed to achieve vibration reduction. And one self-tuning predictive control method is proposed to achieve the stable levitation of this device. Firstly, a predictive control (MPC) method with state constraints and input constraints is adopted to realise the stable suspension of the floater. The MPC can solve the problem of position imbalance of the magnetic levitation system under the external complex disturbances. Secondly, a self-tuning MPC method based on recursive least square is proposed to solve the problem caused by the fixed parameters of the traditional predictive controller. At the beginning of each control cycle, the recursive least-squares (RLS) method is used to estimate the parameters of the system. Thus, the optimal control model could be obtained for the current situation. Then, this model is applied to the predictive controller to solve the problem of parameter fixation in the traditional predictive control. Finally, the simulation results show that it can improve the accuracy, dynamic response and anti-interference performance obviously.

Novel Active Queue Management Scheme for Routers in Wireless Networks

Congestion stability and prevention is the main aim of any decent network traffic management scheme. The protocol that drives congestion control in IP networks, called The Transmission Control Protocol (TCP), is plagued by bandwidth underutilization when implemented on wireless networks. The management of network trunk queue size is the goal of this paper, along with its antecedent delay issues with Active Queue Management (AQM) algorithms that overcomes the bandwidth underutilization issue faced by TCP flows in wireless networks. Two adaptive TCP models, specifically built for wireless networks, were created. This made it possible to implement feedback control techniques in the chosen AQM design for queue management. The two AQM policies were created with the help of Model Predictive Control (MPC) and Self-Tuning Control (STC). A very thorough comparison was made between the developed AQM schemes and two Early Detection Controllers, namely, Random Early Detection (RED) and Fixed-Parameter Proportional Integral (PI). Under varying wireless network conditions, the results obtained from simulating the proposed queue management scheme in MATLAB® confirmed that the fixed PI and RED controllers were, by comparison, inferior in performance. Copyright © 2018 Praise Worthy Prize S.r.l. - All rights reserved.

Self-Tuning Predictive PID Controller Using a Time-Varying Proportional Gain

This paper proposes a new design method of adaptive GPC-based PID controller. The proportional gain of a PID controller is time-varying, and the PID controller is designed based on strongly stable GPC that is extended by using coprime factorization. Hence, a GPC law can be approximated by the PID controller finely. Finally, the system parameters are updated in real-time to resist to disturbance, by using the RLS method. Simulation examples show the effectiveness of the proposed method.

The Application of Implicit Generalized Predictive Self-Tuning Algorithm for Liquefied Natural Gas Storage Station

The advanced implicit generalized predictive self-tuning control algorithm was used against those characteristics of the outlet’s gas pressure such as large-time delay, time varying, vulnerable to random noise and so on. Through the simulation, a good tracking performance was shown, and in the process of actual operation, it also achieved satisfactory control effect.

Self-Tuning Predictive Control of Nonlinear Servo-Motor

Self-Tuning Predictive Control of Nonlinear Servo-MotorThe paper is focused on a design of a self-tuning predictive model control (STMPC) algorithm and its application to a control of a laboratory servo motor. The model predictive control algorithm considers constraints of a manipulated variable. An ARX model is used in the identification part of the self-tuning controller and its parameters are recursively estimated using the recursive least squares method with the directional forgetting. The control algorithm is based on the Generalised Predictive Control (GPC) method and the optimization was realized by minimization of a quadratic and absolute values objective functions. A recursive control algorithm was designed for computation of individual predictions by incorporating a receding horizon principle. Proposed predictive controllers were verified by a real-time control of highly nonlinear laboratory model — Amira DR300.

Design of an Adaptive Self-Tuning Smith Predictor for a Time Varying Water Treatment Process

Abstract Abstract This paper presents the simulation and real time implementation of an adaptive predictive PI controller for the control of Chlorine dosing in secondary water disinfection rigs. This trial is part of a project that looks at the optimisation of process control specifically in the water industry. As one of the main treatment processes, Chlorine dosing is one of the processes that naturally imposes very long dead times for the controller to deal with. Although PI controllers are still commonly used for controlling this process, previous literature as well as trials carried as part of this project proved that the performance of PI controllers, no matter how tuned they are, is very sluggish and unreliable. A pilot rig was used instead of a live secondary disinfection rig, and a number of open loop step tests were performed for system identification. Once the process dynamics became known, complementary functions that estimate the process transfer function based on the water flow were introduced. A standard tuned PI controller configuration was simulated for the process, and then a Smith predictor was used in order to be able to compare the performance of the predictive PI with a standard PI controller. Tuning functions were derived for the PI to make it a self-tuning predictive controller, and parameter estimation functions were also used so that the final outcome is an adaptive self-tuning system. This system was then implemented on the same pilot rig, and real time implementation proved the findings obtained from the simulation. Both simulation and pilot rig tests show a very good dynamic response with excellent accuracy.

Self-tuning control of dead time compensation strategy for PMSM control

A self-tuning predictive control strategy for a permanent magnet synchronous motor (PMSM) drive with dead time is proposed. Firstly, a dead time compensator (DTC), based on the Smith predictor (SP) and a self-tuning (ST) proportional-integral (PI) model following controller (MFC) is proposed. Then the model-following error is used to adaptively update the gains of the MFC via the affine projection algorithm (APA). The control strategy is very simple and easy to be implemented on digital signal processor (DSP). The validity and usefulness of the proposed control scheme are verified through experimental results.

Predictive control of periodic-review production inventory systems with deteriorating items

In this paper a predictive control strategy is applied to a periodic-review dynamic inventory system with deteriorating items. Given the current inventory level, we determine the optimal production rates to be implemented at the beginning of each of the following periods over the control horizon. The effectiveness of this approach is the use of future information of the inventory target level and the desired production rate, which are available, along the fixed horizon. The deterioration coefficient may be known or unknown and both cases are considered. In the case where it is unknown, the self-tuning predictive control is applied. The proposed control algorithms are illustrated by simulations.

A Design of Multiloop Predictive Self-Tuning PID Controllers

In this paper, a design scheme of multiloop predictive self-tuning PID control systems is proposed for multivariable systems. The proposed scheme firstly uses a static pre-compensator as an approximately decoupling device, in order to roughly decouple the controlled object. Furthermore, by regarding the approximately decoupled system as a series of single-input single-output subsystems, a single-input single-output PID controller is designed for each subsystem. The PLD parameters are calculated on-line based on the relationship between the PID control and the generalized predictive control laws.

Model-based control of strip temperature for the heating furnace in continuous annealing

Presents the model of the heating furnace in continuous annealing processes for use in design of self-tuning control systems. A simplified mathematical model is derived from first principles. The model parameters are recursively estimated with an algorithm called recursive parameter estimation with a vector-type variable forgetting factor (REVVF). The REVVF algorithm was developed for such cases where some knowledge on parameter variability can be obtained beforehand. The control system of strip temperature presented here is hierarchical. The upper level is called "optimal preview control", which performs preset control. It previews the approaching setup change, which is the change of strip size or reference temperature, and optimizes the line speed and the strip temperature trajectory. Next, the lower level is called "temperature tracking control", which performs closed-loop control using the above trajectory as the control target. At this level, the generalized pole-placement self-tuning control was first employed; and later, the generalized predictive self-tuning control was introduced. These control methods were applied with some practical modifications and with the above mentioned REVVF. The control has been working successfully in several real plants.

Multivariable self-tuning temperature control for plastic injection molding process

This paper develops a multivariable self-tuning predictive control for improving set-point tracking performance, disturbance rejection, and robustness of a temperature control system for an extruder barrel in a plastic injection molding process. The stochastic discrete-time multivariable mathematical model is built and its unknown system parameters are identified by using the recursive least-squares estimation method. The multivariable predictive control is derived based on the minimization of a generalized predictive performance criterion. A real-time self-tuning control algorithm is proposed and then implemented by using a digital signal processor (DSP) TMS320C31 from Texas Instruments. Experimental results are used to show the feasibility and effectiveness of the proposed method.

A new robust non-linear self-tuning predictive control using neural networks

An adaptive version of a novel robust predictive control for a class of non-linear systems is presented. The non-linear system is separated into linear and non-linear parts by Taylor series expansion and then the latter part is identified by a neural network, which is then compensated in the control algorithm such that feedback linearization can be achieved. Thus the influence of the non-linearity and model uncertainties may be eliminated or reduced. In the case of time-varying or unknown systems the linear part of the system model is estimated by an RLS (recursive least-squares) algorithm. Simulation results show that the proposed scheme may improve the system performance.

Development of an Industrial Self-Tuning and Adaptive Model-Based Predictive Controller

Abstract This paper presents an identification scheme which allows the parametric identification of standard aperiodic minimum-phase processes without any further a priori knowledge of the process. This is achieved by splitting up the identification into two steps. After the first step a temporary PID controller is activated. After the second step the parametric process model obtained is checked with a quality criterion. If the model fits the demands of this criterion it is used to design a model-based predictive controller.

Nonlinear control algorithms and power system application

Two distinct approaches to design of nonlinear control systems are discussed in context of stabilization of interconnected power transmission systems. The control problem considered involves manipulation of line impedance through thyristor controlled series capacitor to increase stability of the transmission system in case of a large fault. Since the control signal affects system dynamics multiplicatively, traditional linear control strategies are found not appropriate and nonlinear techniques must be used instead. The first method considered is a self-tuning predictive controller based on bilinear model structure. The algorithm is shown to quickly identify the system dynamics and effectively damp oscillations following a simulated fault. The second method is a neural-network-based hierarchical controller using simplified models of the system and off-line training to classify and interpolate between different operating conditions of the system. Both nonlinear controllers are based on a concept of optimization of a performance criterion and naturally incorporate limits on control signals.

Self-Tuning Predictive Control with Application to Steam Generator

This paper presents a self-tuning predictive control algorithm for steam generator. The control algorithm is derived by suitably modifying the generalized predictive control algorithm. The main feature of the proposed method is to take the measurable disturbance into problem consideration and a simple adaptive scheme for obtaining the controller gain. In order to evaluate the performance of the proposed algorithm, computer simulations are done for an PWR steam generator model. Simulation results show satisfactory performances against load variations, steam flow rate estimation errors, and feedwater temperature variations. It can be also observed that the proposed algorithm exhibits better responses than a conventional PID controller.

Generalised Pole-Placement Control of Steam Turbine Speed

An application of a pole-placement self-tuning predictive control algorithm is developed to regulate speed of a power plant steam turbine model. Two types of system representation (CARMA and CARIMA) are used to test the control algorithm. Simulation results show that when using a CARMA model produces better results. Two further comparisons are made when using a PI controller and a generalised predictive controller.

Self-tuning predictive control of processing temperature for food extrusion

Abstract A self-tuning regulator was developed for temperature control in a food extrusion process. Process models with on-line identification were used to predict the processing temperature and major disturbances. The regulator was designed in an explicit formulation based on feedforward and internal-model-based predictive control strategies. An extended prediction window was used in the identification and control algorithms for enhanced robustness against uncertainties in time delays. The control system was implemented and tested experimentally. The effectiveness of the algorithms was demonstrated.

Predictive self-tuning control by parameter bounding and worst-case design

The computation of bounds on the parameters of a plant model allows worst-case control synthesis, taking account of the uncertainty in the model. This paper introduces such a control scheme: predictive bounding control. The scheme contrasts with existing self-tuning control methods which base control synthesis on a nominal plant model. Parameter bounding also permits detection of abrupt plant changes and adaptive tracking of time-varying plant characteristics by suitable choice of bounds on plant-model output error and plant-parameter increments. Estimation and control are closely integrated and the control computation can compromise between reducing the model uncertainty and reducing predicted output error. Simulation examples show the excellent performance of predictive bounding control.

THE FEEDBACK REALIZATION OF THE GENERALIZED NYQUIST MULTIVARIABLE PREDICTIVE CONTROL AND ITS APPLICATION

In association with multivariable frequency-design techniques, Cloud and Kouvaritakis (1988) presented a framework for the multivariable generalization of predictive self-tuning control. On the basis of it, this article proposes a closed-feedback structure of multivariable predictive control system projected onto a characteristic frame, from which the information required to assess stability can be obtained. Furthermore, the computational load and the memory storage for implementation can be effectively reduced. In order to verify the validity of the method, the simulation for predictive control algorithm to design the multivariable temperature control system of a 200 MW thermal power unit made by the People's Republic of China is illustrated. The results show that the control system has good control performance.

Generalized predictive self-tuning control for tracking a periodic reference signal

A generalized predictive control scheme for tracking a periodic reference signal, based on the internal model principle, is proposed. Using LQ regulator theory, some stability results are derived. Also, some simulation results for the self-tuning case are presented to see the effectiveness of the proposed method.