Robust Adaptive Backstepping Motion Control of Linear Ultrasonic Motors Using Fuzzy Neural Network

A robust adaptive fuzzy neural network (RAFNN) backstepping control system is proposed to control the position of an X-Y-Theta motion control stage using linear ultrasonic motors (LUSMs) to track various contours in this study. First, an X-Y-Theta motion control stage is introduced. Then, the single-axis dynamics of LUSM mechanism with the introduction of a lumped uncertainty, which includes cross-coupled interference and friction force, is derived. Moreover, a conventional backstepping approach is proposed to compensate the uncertainties occurred in the motion control system. Furthermore, to improve the control performance in the tracking of the reference contours, an RAFNN backstepping control system is proposed to remove the chattering phenomena caused by the sign function in the backstepping control law. In the proposed RAFNN backstepping control system, a Sugeno-type adaptive fuzzy neural network (SAFNN) is employed to estimate the lumped uncertainty directly and a compensator is utilized to confront the reconstructed error of the SAFNN. In addition, the motions at the X axis, Y axis, and Theta axis are controlled separately. The experimental results show that the contour tracking performance is significantly improved and the robustness to parameter variations, external disturbances, cross-coupled interference, and friction force can be obtained, as well using the proposed RAFNN backstepping control system.     

Hybrid controller with recurrent neural network for magnetic levitation system

We propose a hybrid controller using a recurrent neural network (RNN) to control a levitated object in a magnetic levitation system. We describe a nonlinear dynamic model of the system and propose a computed force controller, based on feedback linearization, to control the position of the levitated object. To relax the requirement of the lumped uncertainty in the design of the computed force controller, an RNN functions as an uncertainty observer to adapt the lumped uncertainty on line. The computed force controller, the RNN uncertainty observer, and a compensated controller are embodied in a hybrid controller, which is based on Lyapunov stability. The computed force controller, with the RNN uncertainty observer, is the main tracking controller, and the compensated controller compensates the minimum approximation error of the RNN uncertainty observer. To ensure the convergence of the RNN, the adaptation law of the RNN is modified by using a projection algorithm. Experimental results illustrate the validity of the proposed control design for the magnetic levitation system.     

A fuzzy neural network controller with adaptive learning rates for nonlinear slider-crank mechanism

A fuzzy neural network (FNN) controller with adaptive learning rates is proposed to control a nonlinear mechanism system in this study. First, the network structure and the on-line learning algorithm of the FNN is described. To guarantee the convergence of the tracking error, analytical methods based on a discrete-type Lyapunov function are proposed to determine the adaptive learning rates of the FNN. Next, a slider-crank mechanism, which is driven by a permanent magnet (PM) synchronous motor, is studied as an example to demonstrate the effectiveness of the proposed control technique; the FNN controller is implemented to control the slider position of the motor-slider-crank nonlinear mechanism. The robust control performance and learning ability of the proposed FNN controller with adaptive learning rates is demonstrated by simulation and experimental results.     

Sliding-mode-controlled slider-crank mechanism with fuzzy neuralnetwork

The dynamic response of a sliding-mode-controlled slider-crank mechanism, which is driven by a permanent-magnet (PM) synchronous servo motor, is studied in this paper. First, a position controller is developed based on the principles of sliding-mode control. Moreover, to relax the requirement of the bound of uncertainties in the design of a sliding-mode controller, a fuzzy neural network (FNN) sliding-mode controller is investigated, in which a FNN is adopted to adjust the control gain in a switching control law on line to satisfy the sliding mode condition. In addition, to guarantee the convergence of tracking error, analytical methods based on a discrete-type Lyapunov function are proposed to determine the varied learning rates of the FNN. Numerical and experimental results show that the dynamic behaviors of the proposed controller-motor-mechanism system are robust with regard to parametric variations and external disturbances. Furthermore, compared with the sliding-mode controller, smaller control effort results and the chattering phenomenon is much reduced by the proposed FNN sliding-mode controller     

Decentralized PID neural network control for five degree-of-freedom active magneticbearing

A decentralized proportional–integral–derivative neural network (PIDNN) control scheme is proposed to regulate and stabilize a fully suspended five degree-of-freedom (DOF) active magnetic bearing (AMB) system which is composed of two radial AMBs (RAMBs) and one thrust AMB (TAMB). First, the structure and operating principles of the five-DOF AMB system are introduced. Then, the adopted differential driving mode (DDM) for the drive system of the AMB is analyzed. Moreover, due to the exact dynamic model of the five-DOF AMB system is vague, a decentralized PIDNN controller is proposed to control the five-axes of the rotor simultaneously in order to confront the uncertainties including inherent nonlinearities and external disturbances effectively. Furthermore, the connective weights of the PIDNN are trained on-line and the convergence analysis of the regulating error is provided using a discrete-type Lyapunov function. Based on the decentralized concepts, the computational burden is reduced and the controller design is simplified. Finally, the experimental results show that the proposed control scheme provides good control performances and robustness for controlling the fully suspended five-DOF AMB system in different operating conditions.     

A permanent-magnet synchronous motor servo drive using self-constructing fuzzy neural network controller

A self-constructing fuzzy neural network (SCFNN) is proposed to control the rotor position of a permanent-magnet synchronous motor (PMSM) drive to track periodic step and sinusoidal reference inputs in this study. The structure and the parameter learning phases are preformed concurrently and online in the SCFNN. The structure learning is based on the partition of input space, and the parameter learning is based on the supervised gradient descent method using a delta adaptation law. Several simulation and experimental results are provided to demonstrate the effectiveness of the proposed SCFNN control stratagem under the occurrence of parameter variations and external disturbance.     

Recurrent Functional-Link-Based Fuzzy-Neural-Network-Controlled Induction-Generator System Using Improved Particle Swarm Optimization

A recurrent functional-link (FL)-based fuzzy-neural-network (FNN) controller with improved particle swarm optimization (IPSO) is proposed in this paper to control a three-phase induction-generator (IG) system for stand-alone power application. First, an indirect field-oriented mechanism is implemented for the control of the IG. Then, an AC/DC power converter and a DC/AC power inverter are developed to convert the electric power generated by a three-phase IG from variable frequency and variable voltage to constant frequency and constant voltage, respectively. Moreover, two online-trained recurrent FL-based FNNs are introduced as the regulating controllers for both the DC-link voltage of the AC/DC power converter and the AC line voltage of the DC/AC power inverter. Furthermore, IPSO is adopted to adjust the learning rates to improve the online learning capability of the recurrent FL-based FNNs. Finally, some experimental results are provided to demonstrate the effectiveness of the proposed recurrent FL-based FNN-controlled IG system.     

Intelligent controlled shunt active power filter for voltage and current harmonic compensation in microgrid system

An integrated microgrid with a novel shunt active power filter (APF) using Elman neural network (ENN) is proposed in this study. The microgrid consists of a storage system, a photovoltaic (PV) system, the shunt APF, a linear load, and a nonlinear load. Moreover, the master/slave control algorithm is adopted in the microgrid. The storage system, which is considered as the master unit, is adopted to control the active and reactive power outputs (P/Q control) in grid-connected mode and the voltage and frequency of the microgrid (V/f control) in islanded mode. Furthermore, the PV system is considered as the slave unit to implement P/Q control in both grid-connected and islanded modes. In addition, the proposed shunt APF possesses dual functions of voltage and current harmonic compensation for microgrid under voltage harmonic propagation and nonlinear load to reduce the voltage and current total harmonic distortions (THD) effectively. Additionally, an ENN controller is adopted in the proposed shunt APF to improve the transient and steady-state responses of DC-link voltage during the switching between the grid-connected mode and islanded mode. Finally, some simulation results are provided to verify the feasibility and the effectiveness of the integrated microgrid with the intelligent controlled shunt APF.     

An adaptive recurrent radial basis function network tracking controller for a two-dimensional piezo-positioning stage

An adaptive recurrent radial basis function network (ARRBFN) tracking controller for a two-dimensional piezo-positioning stage is proposed in this study. First, a mathematical model that represents the dynamics of the two-dimensional piezo-positioning stage is proposed. In this model, a hysteresis friction force that describes the hysteresis behavior of one-dimensional motion is used; and a nonconstant stiffness with the cross-coupling dynamic due to the effect of bending of a lever mechanism in x and y axes also is included. Then, according to the proposed mathematical model, an ARRBFN tracking controller is proposed. In the proposed ARRBFN control system, a recurrent radial basis function network (RRBFN) with accurate approximation capability is used to approximate an unknown dynamic function. The adaptive learning algorithms that can learn the parameters of the RRBFN on line are derived using Lyapunov stability theorem. Moreover, a robust compensator is proposed to confront the uncertainties, including approximation error, optimal parameter vectors, higher-order terms in Taylor series. To relax the requirement of the value of the lumped uncertainty in the robust compensator, an adaptive law is investigated to estimate the lumped uncertainty. Using the proposed control scheme, the position tracking performance is substantially improved and the robustness to uncertainties, including hysteresis friction force and cross-coupling stiffness, can be obtained as well. The tracking performance and the robustness to external load of the proposed ARRBFN control system are illustrated by some experimental results.     

Recurrent RBFN-based fuzzy neural network control for X-Y-theta motion control stage using linear ultrasonic motors

A recurrent radial basis function network (RBFN) based fuzzy neural network (FNN) control system is proposed to control the position of an X-Y-theta motion control stage using linear ultrasonic motors (LUSMs) to track various contours in this study. The proposed recurrent RBFN-based FNN combines the merits of self-constructing fuzzy neural network (SCFNN), recurrent neural network (RNN), and RBFN. Moreover, the structure and the parameter learning phases of the recurrent RBFN-based FNN are performed concurrently and on line. The structure learning is based on the partition of input space, and the parameter learning is based on the supervised gradient decent method using a delta adaptation law. The experimental results due to various contours show that the dynamic behaviors of the proposed recurrent RBFN-based FNN control system are robust with regard to uncertainties.