FUZZY PREDICTIVE CONTROL OF UNCERTAIN CHAOTIC SYSTEMS USING TIME SERIES

1999 ◽  
Vol 09 (04) ◽  
pp. 757-767 ◽  
Author(s):  
LIANG CHEN ◽  
GUANRONG CHEN
Keyword(s):  
Chaotic System ◽  
Controller Design ◽  
Recursive Least Squares ◽  
Input Output ◽  
Output Data ◽  
Lyapunov Stability Criterion ◽  
System Input ◽  
Prediction And Control ◽  
And Control ◽  
Fuzzy Predictive Control

In this paper, a simple fuzzy logic based intelligent mechanism is developed for predicting and controlling a chaotic system to a desired target, using only input–output data obtained from the unknown (or uncertain) underlying chaotic system. In the chaos prediction phase, a fuzzy system approach incorporating with Gaussian type of fuzzy membership functions is used. Only system input–output data are needed for prediction, and a recursive least-squares computational algorithm is employed for the calculation. In the controller design phase, the Lyapunov stability criterion is used, which forms the basis of the main design principle. Some simulation results on the chaotic Sin map and Hénon map are given, for both prediction and control, to illustrate the effectiveness and control performance of the proposed method.

Model-Matching Controller Design With Input-Output Data—A Numerical Approach for Redundantly Actuated Systems

1994 ◽  
Vol 116 (4) ◽  
pp. 800-805
Author(s):  
Jenq-Tzong H. Chan
Keyword(s):  
Explicit Knowledge ◽  
Controller Design ◽  
Numerical Technique ◽  
Open Loop ◽  
Loop System ◽  
Synthesis Process ◽  
Model Matching ◽  
Input Output ◽  
Output Data ◽  
Redundantly Actuated

A numerical technique for control system synthesis based on input-output data is presented. The method is applicable when the system is open-loop stable and redundantly actuated. The major merits of the method are as follows. First, the closed-loop system equation may be arbitrarily assigned. Second, explicit knowledge of an open-loop system model is not needed for controller synthesis. Third, the stability of the synthesized system may be verified during the synthesis process; hence, the workability of the controller is ensured.

scholarly journals Tuning heuristics and convergence analysis of reinforcement learning algorithm for online data-based optimal control design

2020 ◽  
Vol 9 (2) ◽  
pp. e188922128
Author(s):  
Fábio Nogueira da Silva ◽  
João Viana Fonseca Neto
Keyword(s):  
Reinforcement Learning ◽  
Convergence Analysis ◽  
Output Feedback ◽  
Learning Algorithm ◽  
Policy Iteration ◽  
Recursive Least Squares ◽  
Input Output ◽  
Output Data ◽  
Data Generator ◽  
Reinforcement Learning Algorithm

A heuristic for tuning and convergence analysis of the reinforcement learning algorithm for control with output feedback with only input / output data generated by a model is presented. To promote convergence analysis, it is necessary to perform the parameter adjustment in the algorithms used for data generation, and iteratively solve the control problem. A heuristic is proposed to adjust the data generator parameters creating surfaces to assist in the convergence and robustness analysis process of the optimal online control methodology. The algorithm tested is the discrete linear quadratic regulator (DLQR) with output feedback, based on reinforcement learning algorithms through temporal difference learning in the policy iteration scheme to determine the optimal policy using input / output data only. In the policy iteration algorithm, recursive least squares (RLS) is used to estimate online parameters associated with output feedback DLQR. After applying the proposed tuning heuristics, the influence of the parameters could be clearly seen, and the convergence analysis facilitated.

A Modified Recursive Identifier Not Requiring Initial Condition Estimates

1982 ◽  
Vol 104 (3) ◽  
pp. 264-266
Author(s):  
J. M. Mocenigo ◽  
A. E. Pearson
Keyword(s):  
Least Squares ◽  
Initial Conditions ◽  
Least Squares Estimator ◽  
Recursive Least Squares ◽  
Stochastic Noise ◽  
Initial Condition ◽  
Limited Data ◽  
Input Output ◽  
Output Data

A recursive least-squares estimator is developed in an identification observer format with the property that the need for estimating initial conditions is eliminated for time limited data. Estimators are developed that work with observed input/output data corrupted by deterministic, piecewise deterministic, or stochastic noise.

scholarly journals Detection of Vibrations Defects in Gas Transportation Plant Based on Input / Output Data Analysis: Gas Turbine Investigations

2020 ◽  
Vol 25 (4) ◽  
pp. 42-58
Author(s):  
B. Djaidir ◽  
A. Hafaifa ◽  
M. Guemana ◽  
A Kouzou
Keyword(s):  
Gas Turbine ◽  
Gas Turbines ◽  
Oil And Gas ◽  
Input Output ◽  
Output Data ◽  
Fault Monitoring ◽  
Main Components ◽  
And Control ◽  
Detection And Localization ◽  
Proper Operation

AbstractIn oil and gas industrial production and transportation plants, gas turbines are considered to be the major pieces of equipment exposed to several unstable phenomena presenting a serious danger to their proper operation and to their exploitation. The main objective of this work is to improve the competitiveness performance of this type of equipment by analyses and control of the dynamic behaviors and to develop a fault monitoring system for the equipment exposed and subject to certain eventual anomalies related to the main components, namely the shaft and the rotors. This study will allow the detection and localization of vibration phenomena in the studied gas turbine based on the input / output data.

Sign in / Sign up

Export Citation Format

Share Document