Gu Shusheng

Fuzzy adaptive output feedback control for nonlinear MIMO systems based on observer

Fuzzy adaptive control method based on observer is proposed for a class of MIMO nonlinear systems. In the design process, the approximation capability of the fuzzy logic system is used, a kind of high-gain observer is introduced, and the output feedback control law and the parameter adaptive law are derived by using Lyapunov method. This control method considers the existence of approximation error and the systems external disturbance, which have an unknown boundary and need not the assumption of measurements of states. It is proved that the tracking performance of the output feedback control can achieve that of the state feedback control.

A neural-network-base adaptive estimator for speed-sensorless control of induction motor

This paper presents an adaptive estimator for speed-sensorless field-oriented control of induction motor. By measuring the phase voltages and currents induction motor (IM) drive, a neural network based rotor flux components and speed estimation method for IM is described. The proposed estimator includes two recurrent neural networks (RNN), one is used to estimate rotor flux and speed, the other is used to estimate stator current. Using an improved recursive prediction error algorithm, online adaptive estimation is realized. The simulation results show high accuracy of the estimation algorithm, and verify the usefulness algorithm.

PID-like controller using a modified neural network

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.Abstract This paper presents an adaptive PID-like controller (PIDLC) using a modified Neural network (MNN) for learning the characteristics of a dynamic system. A PID-like cost function is proposed, and it can be shown that the Proportional-Integral-Dijferential (PID) algorithm is the gradient descent method if a specific cost function is selected. The PIDLC can cope with parametric variation and uncertainty in the controlled plant through online learning. The MNNs learning algorithm is considerably faster through the introduction of a recursive least squares (RLS) algorithm. A simulation shows that this kind of control algorithm is very effective, especially when there are variations in the plant dynamics.

Identification and control of induction motor using artificial neural networks

This paper proposes the use of one kind of artificial neural network (ANN), so called dynamic recurrent neural networks (DRNN), to identify and control an induction motor. A scheme of identification of the electrical dynamics of a voltage-fed induction motor is presented. Computer simulation results of rotor speed are based on a system to adaptively control the rotor speed via identification of the mechanical and current-fed system dynamics, and the performance of the combined speed and current control scheme is shown to be perfect.

A neural-network-based adaptive estimator of rotor position and speed for permanent magnet synchronous motor

In this paper, by measuring the phase voltages and currents of the permanent magnet synchronous motor (PMSM) drive, a neural-network-based rotor position and speed estimation method for PMSM is described. The proposed estimator includes two recurrent neural networks, one is used to estimate rotor speed and rotor position, and the other is used to estimate stator current. Through using an improved recursive prediction error algorithm, on-line adaptative estimation is realized. The simulation results show that the proposed approach gives a good estimation of rotor speed and position. Especially, the proposed approach has low sensitivity to perturbations of the mechanical parameters and torque disturbances.

An Adaptive Filter Model Based on Wavelet Transform

Due to the fact that it is not easy to filter out the overlap noise between noisy signal and noise using the traditional method of wavelet denoising, an adaptive filter model based on the wavelet transform is constructed. In this model, the adaptive filter is used to filter out noise secondary on the basis of first wavelet denoising on the original noisy signal. The experimental results show that the model can effectively remove the noise. And comparing with the original method of wavelet denosing, the SNR of the signal which uses the model to filter out noise is much higher, reaching a better filtering effect.

Robust non-fragile state feedback control of discrete time-delay systems

The problem of designing robust non-fragile state feedback controllers for linear discrete systems with uncertainties and constant delays is studied. The system uncertainties are assumed to be polytopic, and the state feedback gain uncertainties are norm-bounded. The H/sub /spl infin// norm is used to describe system performance. A method of designing robust non-fragile H/sub /spl infin// controllers is proposed in terms of linear matrix inequalities (LMIs). Numerical examples are given to illustrate the proposed results.

An improved recursive prediction error algorithm for training recurrent neural networks

In this paper, a fast and effective learning algorithm for training recurrent neural networks, which is realized by introducing and improving the recursive prediction error (RPE) method, is proposed. The improving scheme for RPE algorithm is adding a momentum term in the gradient of Gauss-Newton search direction and using the changeable forgetting factor. Simulation results show that the proposed algorithm achieves far better convergence performance than the classical backpropagation with the momentum term algorithm, and has superior performance compared with the conventional RPE algorithm.

Robust model predictive control for uncertain discrete-time system with both states and input delays

A novel robust model predictive control strategy is presented for a class of discrete-time time-varying system with delays. The modified quadratic Lyapunov-Krasovskii functionals for uncertain discrete-time system with both states and input delays is proposed. A robust model predictive control law is derived such that the closed-loop system is asymptotically stable and a dasiaworst-casepsila infinite horizon performance objective is minimized. A feasible state feedback model predictive control law is obtained by using linear matrix inequalities approach. And then, a convex optimization problem is formulated to design the robust model predictive controller which minimizes the upper bound of the closed-loop performance objective. The robust model predictive controller design is illustrated with numerical examples.

Knowledge based pipeline network peeling off recognition method of maps

Map recognition is an essential data input means of Geographic Information System (GIS). How to solve the problems in the procedure, such as the recognition of maps with crisscross pipeline networks, the classification of buildings and roads, and the processing of connected text, is a critical step for GIS keeping high-speed development. In this paper, a new recognition method of pipeline maps is presented, and some common patterns of pipeline connection and component labels are established. Through pattern matching, pipelines and component labels are recognized and peeled off from maps. After this approach, maps simply consist of buildings and roads, which are recognized and classified with fuzzy classification method in the next approach. In addition, the Double Sides Scan (DSS) technique is also described, through which the effect of connected text can be eliminated.