Neural Network Based Controllers

An improved Direct Torque Controlled (DTC) Induction Motor (IM) is reported in this paper with the aims to produce an adaptive flux controller design to realize the maximum efficiency in DTC IM drives. The value of reference flux is identified through the artificial intelligent neural network (ANN) algorithm with the input power as the objective function. The description of neural network control system as well as the training procedure is explained in this paper. Consequently, the proposed efficient optimizing controller yields an adaptive reference flux, which ensures a minimum input power that leading to the maximum efficiency of the drives systems is achieved. The proposed schemes have been developed and the performance of the IM Drive under different operating condition has been investigated through simulation and experimentally by using the Simulink/MATLAB and digital signal processor of dSPACE. The promising results validate the effectiveness.

A space vector modulation based direct torque Control strategy is suggested and an intelligence controller design based on this strategy is presented. A neural network controller is proposed to replace the conventional PID controllers to improve the drive's performance since the performance of an electric drive really depends on the quality of a speed controller. The neural network controller was trained and realizes for a speed controller. The controller was utilized in the feedback loop of the control system. The control system renditions as well as the online learning technique of the neural network are described in this paper. The comparison with the conventional PID direct torque controller reveal the effectiveness of the proposed scheme by improved the performance of transient response is presented. A simulation model representing the complete neural network based direct torque control scheme of induction motor drive is developed and verified using S-Function Simulink block program.

The beam wander on the detector plane is one of the main causes for major power loss which severely degrades the performance of Free Space Optical (FSO) link. Designing a suitable con troller to correct the beam motion at a faster rate to increase the beam stability becomes significant. This paper presents an investigation on the performance of two types of controller designed for aiming a laser beam to be at a particular point under dynamic disturbances. The first design is based on the Taguchi's method (direct controller) while the second is the Artificial Neural Network (ANN) method (neural controller). These controllers process the beam location information and generate the neces sary outputs to mitigate the beam wandering. The pipelined parallel architecture for both controllers are developed in a Field Programmable Gate Array (FPGA) and installed at the receiver station. Evi dence of the suitability and the effectiveness of the proposed controller in terms of prediction exact ness, prediction error, reduction of focal point wander, response to impulse and scintillation are pro vided through experimental results from the FSO link established for the horizontal range of 0.5 km at an altitude of 15.25 m.

This paper deal with the problem in speed controller for Indirect Field Oriented Control of Induction Motor. The problem cause decrease performance of Induction Motor where it widely used in high-performance applications. In order decrease the fault of speed induction motor, Takagi-Sugeno type Fuzzy logic control is used as the speed controller. For this, a model of indirect field oriented control of induction motor is built and simulating using MATLAB simulink. Secondly, error of speed and derivative error as the input and change of torque command as the output for speed control is applied in simulation. Lastly, from the simulation result overshoot is zero persent, rise time is 0.4s and settling time is 0.4s. The important data is steady state error is 0.01 percent show that the speed can follow reference speed. From that simulation result illustrate the effectiveness of the proposed approach.

This paper deal with the problem in speed controller for Indirect Field Oriented Control of Induction Motor. The problem cause decrease performance of Induction Motor where it widely used in high-performance applications. In order decrease the fault of speed induction motor, Takagi-Sugeno type Fuzzy logic control is used as the speed controller. For this, a model of indirect field oriented control of induction motor is built and simulating using MATLAB simulink. Secondly, error of speed and derivative error as the input and change of torque command as the output for speed control is applied in simulation. Lastly, from the simulation result overshoot is zero persent, rise time is 0.4s and settling time is 0.4s. The important data is steady state error is 0.01 percent show that the speed can follow reference speed. From that simulation result illustrate the effectiveness of the proposed approach.

A space vector modulation based direct torque Control strategy is suggested and an intelligence controller design based on this strategy is presented. A neural network controller is proposed to replace the conventional PID controllers to improve the drive's performance since the performance of an electric drive really depends on the quality of a speed controller. The neural network controller was trained and realizes for a speed controller. The controller was utilized in the feedback loop of the control system. The control system renditions as well as the online learning technique of the neural network are described in this paper. The comparison with the conventional PID direct torque controller reveal the effectiveness of the proposed scheme by improved the performance of transient response is presented. A simulation model representing the complete neural network based direct torque control scheme of induction motor drive is developed and verified using S-Function Simulink block program.

Development of a neural network (NN) based indirect field oriented control (IFOC) of induction motor (IM) drive is presented in this paper. The neural network indirect field oriented control (NNIFOC) is designed to generate signal of quadrature current, hence the motor speed will follow the reference speed value. The simulation model using Matlab/Simulink was developed to validate the effectiveness of the proposed controller. The simulation results show that the proposed NNIFOC schemes of IM drive is more effective and found to be a proper replacement of the proportional integral derivative (PID) controller.

This paper proposed improvement performance from offline towards online neural network scheme for speed control of induction motor field oriented control based on load disturbance and parameter variation. The neural network is design as 1-3-1 network structure by feedforward architecture to maintain the speed trajectory specified by reference model. Both offline and online networks were trained by backpropagation algorithm and the updating of weight and bias will be done in the online model. Simulation model were developed for both networks by using MATLAB/Simulink software and the results shows that the performance of online NNIFOC improved rather than offline NNIFOC and robust to load disturbance and parameter variation. Key-Words: Neural Network, Induction Motor, Field Oriented Control

Electric  motor  drive  systems  (EMDS)  have  been  recognized  as  one  of the  most  promising  motor  systems  recently  due  to  their  low  energy consumption and reduced emissions. With only some exceptions, EMDS are the  main  source  for  the  provision  of  mechanical  energy  in  industry  and accounts  for  about  60%  of  global  industrial  electricity  consumption.  Large energy  efficiency  potentials  have  been  identified  in  EMDS  with  very  short payback time and high-cost effectiveness. Typical, during operation at rated mode,  the motor  drive  able  to  hold  its  good  efficiencies.  However,  a  motor usually operates out from rated mode in many applications, especially while under  light  load,  it  reduced  the  motor’s  efficiency  severely.  Hence,  it  is necessary that a conventional drive system to embed with loss minimization strategy  to  optimize  the  drive  system  efficiency  over  all  operation  range. Conventionally,  the  flux  value  is  keeping  constantly  over  the  range  of operation,  where  it  should  be  highlighted  that  for  any  operating  point, the losses could be minimize with the proper adjustment of the flux level to a suitable value at that point. Hence, with the intention to generate an adaptive flux  level  corresponding  to  any  operating  point,  especially  at  light  load condition,  an  online  learning  Artificial  Neural  Network  (ANN)  controller was  proposed  in  this  study,  to  minimize  the  system  losses.  The  entire proposed      strategic      drive      system      would      be      verified      under the  MATLAB/Simulink  software  environment.  It  is  expected  that  with the proposed online learning Artificial Neural Network controller efficiency optimization  algorithm  can  achieve  better  energy  saving  compared  with traditional blended strategies.

This paper describes comparative study of various controllers on Rotary Inverted Pendulum (RIP). PID, LQR, FUZZY LOGIC and H∞ controllers are tried on RIP in MatLab Simulink. The same four controllers have been tested on test bed of RIP system the controllers are compared from various aspects. The controllers in simulink are compared with the controllers in real time.

This paper deal with the problem in speed controller for Indirect Field Oriented Control of Induction Motor. The problem cause decrease performance of Induction Motor where it widely used in high-performance applications. In order decrease the fault of speed induction motor, Takagi-Sugeno type Fuzzy logic control is used as the speed controller. For this, a model of indirect field oriented control of induction motor is built and simulating using MATLAB simulink. Secondly, error of speed and derivative error as the input and change of torque command as the output for speed control is applied in simulation. Lastly, from the simulation result overshoot is zero persent, rise time is 0.4s and settling time is 0.4s. The important data is steady state error is 0.01 percent show that the speed can follow reference speed. From that simulation result illustrate the effectiveness of the proposed approach.

A space vector modulation based direct torque Control strategy is suggested and an intelligence controller design based on this strategy is presented. A neural network controller is proposed to replace the conventional PID controllers to improve the drive's performance since the performance of an electric drive really depends on the quality of a speed controller. The neural network controller was trained and realizes for a speed controller. The controller was utilized in the feedback loop of the control system. The control system renditions as well as the online learning technique of the neural network are described in this paper. The comparison with the conventional PID direct torque controller reveal the effectiveness of the proposed scheme by improved the performance of transient response is presented. A simulation model representing the complete neural network based direct torque control scheme of induction motor drive is developed and verified using S-Function Simulink block program.

The field oriented control theory and space vector pulse width modulation technique make a permanent magnet synchronous motor can achieve the performance as well as a DC motor. However, due to the nonlinearity of the permanent magnet synchronous motor drive characteristics, it is difficult to control by using conventional proportional-integral-derivative controller. By this reason in this paper an online neural network controller for the permanent magnet synchronous motor is proposed. The controller is designed to tracks variations of speed references and also during load disturbance. The effectiveness of the proposed method is verified by develop simulation model in MATLAB-simulink program. The simulation results show that the proposed controller can reduce the overshoot, settling time and rise time. It can be concluded that the performance of the controller is improved.

Neural network based indirect field oriented control of induction motor drive is presented in this paper. The purpose of this motor is to investigate the performance of neural network controller which is applied on three phase induction motor drive system. The introduction of neural network using indirect field oriented control principles for decoupling control of induction motor are discusses. A decoupling controller and flux estimator were implemented upon this neural network using MATLAB/SIMULINK software. The training procedure of the neural network, description of the control system is given in this paper. The comparative results show that the neural network based indirect field oriented control schemes of induction motor drive is more effective and, hence, found to be a suitable replacement of the conventional proportional integral derivative controller.

This paper employs a strategy based on Hardware-in-the-Loop (HIL) concept to control 3-phase induction motor on-line. The strategy in this work is divided into 3 stages: on-line identification, off-line controller design, and on-line control. The scope of this paper is limited to the controller part and therefore addresses the last two stages of the strategy. The implementation of the strategy was accomplished using HIL within a Simulink environment. Experimental results for different speed profiles are presented which show the validity of the proposed strategy. Advantages include optimizing time resources and minimizing the cost. This is a result of designing the controller offline and applying different virtual controllers to the system.

This paper proposed improvement performance from offline towards online neural network scheme for speed control of induction motor field oriented control based on load disturbance and parameter variation. The neural network is design as 1-3-1 network structure by feedforward architecture to maintain the speed trajectory specified by reference model. Both offline and online networks were trained by backpropagation algorithm and the updating of weight and bias will be done in the online model. Simulation model were developed for both networks by using MATLAB/Simulink software and the results shows that the performance of online NNIFOC improved rather than offline NNIFOC and robust to load disturbance and parameter variation.

A space vector modulation based direct torque Control strategy is suggested and an intelligence controller design based on this strategy is presented. A neural network controller is proposed to replace the conventional PID controllers to improve the drive’s performance since the performance of an electric drive really depends on the quality of a speed controller. The neural network controller was trained and realizes for a speed controller. The controller was utilized in the feed-back loop of the control system. The control system renditions as well as the online learning technique of the neural network are described in this paper. The comparison with the conventional PID direct torque controller reveal the effectiveness of the proposed scheme by improved the performance of transient response is presented. A simulation model representing the complete neural network based direct torque control scheme of induction motor drive is developed and verified using S-Function Simulink block program.

Direct Torque Control (DTC) of Induction Motor drive has quick torque response without complex orientation transformation and inner loop current control. Although DTC has some drawbacks, such as the torque and flux ripple. The important point in DTC is the right selection of the stator voltage vector.This paper presents simple structured neural networks for flux position estimation, sector selection and stator voltage vector selection for induction motors using direct torque control (DTC) method. The Levenberg-Marquardt back-propagation technique has been used to train the neural network. The simple structure network facilitates a short training and processing times. The conventional flux position estimator, sector selector and stator voltage vector selector based DTC scheme compared with the proposed scheme and the results are validated through simulation