Artificial intelligence (AI) techniques, particularly the neural networks, are recently having significant impact on power electronics and motor drives. Neural networks have created a new and advancing frontier in power electronics, which is already a complex and multidisciplinary technology that is going through dynamic evolution in the recent years. This paper gives a comprehensive introduction and perspective of neural network applications in the intelligent control and estimation for power electronics and motor drives area. The principal topologies of neural networks that are currently most relevant for applications in power electronics have been reviewed including the detailed description of their properties. Both feedforward and feedback or recurrent architectures have been covered in the description. The application examples that are discussed in this paper include nonlinear function generation, delayless filtering and waveform processing, feedback signal processing of vector drive, space vector PWM of two-level and multilevel inverters, adaptive flux vector estimation, and some of their combination for vector-controlled ac drive. Additional selected applications in the literature are included in the references. From the current trend of the technology, it appears that neural networks will find widespread applications in power electronics and motor drives in future

Neural Network Applications in Power Electronics and Motor Drives—An Introduction and Perspective

This paper presents a novel speed control scheme of an induction motor (IM) using fuzzy logic control. The fuzzy logic controller (FLC) is based on the indirect vector control. The fuzzy logic speed controller is employed in the outer loop. The complete vector control scheme of the IM drive incorporating the FLC is experimentally implemented using a digital signal processor board DS-1102 for the laboratory 1 hp squirrel cage induction motor. The performances of the proposed FLC based IM drive are investigated and compared to those obtained from the conventional proportional integral (PI) controller based drive both theoretically and experimentally at different dynamic operating conditions such as sudden change in command speed, step change in load, etc. The comparative experimental results show that the FLC is more robust and hence found to be a suitable replacement of the conventional PI controller for the high performance industrial drive applications.

Performances of novel fuzzy logic based indirect vector control for induction motor drive

Description: An advanced introduction to the simulation and hardware implementation of BLDC motor drives A thorough reference on the simulation and hardware implementation of BLDC motor drives, this book covers recent advances in the control of BLDC motor drives, including intelligent control, sensorless control, torque ripple reduction and hardware implementation. With the guidance of the expert author team, readers will understand the principle, modelling, design and control of BLDC motor drives. The advanced control methods and new achievements of BLDC motor drives, of interest to more advanced readers, are also presented. Focuses on the control of PM brushless DC motors, giving readers the foundations to the topic that they can build on through more advanced reading Systematically guides readers through the subject, introducing basic operational principles before moving on to advanced control algorithms and implementations Covers special issues, such as sensorless control, intelligent control, torque ripple reduction and hardware implementation, which also have applications to other types of motors Includes presentation files with lecture notes and Matlab 7 coding on a companion website for the book

Permanent Magnet Brushless DC Motor Drives and Controls: Xia/Permanent Magnet Brushless DC Motor Drives and Controls

Vector control is one of the standard techniques used for the control of a permanent magnet synchronous motor (PMSM). The outer speed loop in vector controlled PMSM drive greatly affects the drive performance. In order to combine the advantages of proportional plus integral (PI) and fuzzy controllers, hybrid fuzzy-PI controllers are used in which the output can either be the outputs of the two, i.e. the PI or fuzzy units being switched as per the predetermined speed errors or be a combination of the two outputs with separate weights assigned to them with online calculations for the weights from the speed errors. The former method based on switching often causes chattering effects, and later method demands larger execution time because of inclusion of separate switching algorithms. This paper reports the vector control of PMSM with hybrid fuzzy-PI speed controller with switching functions calculated based on the weights for both the controller outputs using the output of (a) only the fuzzy controller, (b) only the PI controller and (c) a combination of the outputs of both the controllers. These switching functions are very simple and effective and do not demand any extra computations to arrive at the hybrid fuzzy-PI controller outputs. These control algorithms have been simulated and also implemented on hardware with TMS320F2812 digital signal processor, and it is observed that the performance of the vector controlled PMSM drive with these hybrid fuzzy-PI speed controllers in terms of the response and torque ripples is very promising.

PM Synchronous Motor Speed Control Using Hybrid Fuzzy-PI With Novel Switching Functions

This paper presents an improved fuzzy logic controller (FLC) for an interior permanent magnet synchronous motor (IPMSM) for high-performance industrial drive applications. In the proposed control scheme for high-speed operations above the rated speed, the operating limits of IPMSM are expanded by incorporating the maximum torque per ampere operation in constant torque region and the flux-weakening operation in constant power region. The power ratings of the motor and the inverter are considered in developing the control algorithm. A new and simple FLC is utilized as a speed controller. The FLC is developed to have less computational burden, which makes it suitable for real-time implementation, particularly at high-speed operating conditions. The complete drive is implemented in real-time using digital signal processor (DSP) controller board DS 1102 on a laboratory 1-hp IPM motor. The efficiency of the proposed control scheme is evaluated through both experimental and computer simulation results. The proposed controller is found to be robust for high-speed applications

High-Speed Control of IPMSM Drives Using Improved Fuzzy Logic Algorithms

A brushless DC motor drive with a proposed adaptive fuzzy-neural-network controller is introduced in this paper. First, a neural network-based architecture is introduced for fuzzy logic control. The characteristic rules and their membership functions of fuzzy systems are represented as the processing nodes in the neural network structure. Then, the fuzzy rules and input/output of the system are tuned by the supervised gradient descent learning algorithm. Using an experimental setup, the performance of the proposed controller is evaluated under various operating conditions. Test results are presented and discussed in the paper. The presented controller is shown to be robust, adaptive and capable of learning. To demonstrate the effectiveness of the controller, a proportional-integral controller has been used to perform comparative studies with encouraging results.

Development and implementation of an adaptive fuzzy-neural-network controller for brushless drives

In this paper, a high-performance controller with simultaneous online identification and control is designed for brushless DC motor drives. The dynamics of the motor/load are modeled "online" and controlled using two different neural network based identification and control schemes, as the system is in operation. In the first scheme, an attempt is made to control the rotor angular speed, utilizing a single three-hidden-layer network. The second scheme attempts to control the stator currents, using a predetermined control law as a function of the estimated states. This scheme incorporates three multilayered feedforward neural networks that the online trained, using the Levenburg-Marquadt training algorithm. The control of the direct and quadrature components of the stator current successfully tracked a wide variety of trajectories after relatively short online training periods. The control strategy adapts to the uncertainties of the motor/load dynamics and, in addition, learns their inherent nonlinearities. Simulation results illustrated that a neurocontroller used in conjunction with adaptive control schemes can result in a flexible control device which may be utilized in a wide range of environments.

A continually online-trained neural network controller for brushless DC motor drives

This paper describes the design and experimental verification of a hybrid fuzzy control system for a high-performance brushless DC motor drive. Both the design of the fuzzy controller and its integration with the proportional-integral (PI) in a global control system are discussed. The principle, of the proposed control system is to use a PI controller, which performs satisfactorily in most cases, while keeping in the background, a fuzzy controller, which is ready to take over the PI controller when severe perturbations occur. Strategies of the switching control for the hybrid controller is such that the fuzzy controller is activated whenever the resident system exhibits an oscillatory and/or overshoot behavior. Thus, the PI and fuzzy controllers can be managed to take advantage of their positive attributes. Performance of the hybrid fuzzy-PI controller is evaluated through a laboratory implementation. The laboratory implementation is based on a linguistic fuzzy controller whose design is derived from the expert knowledge gained during disturbances. Experimental results have shown excellent tracking performance of the proposed control system, and have convincingly demonstrated the usefulness of the hybrid fuzzy controller in high-performance drives with uncertainties.

Experimental verification of a hybrid fuzzy control strategy for a high-performance brushless DC drive system

This paper presents an adaptive parallel control architecture, using an artificial neural network (ANN) which is trained while the controller is operating online. The proposed control structure incorporates five-multilayer feedforward ANNs, which are online trained using the Marquardt-Levenberg least-squares learning algorithm. The five networks are used exclusively for system estimation. The estimation mechanism uses online training to learn the unknown model dynamics and estimate the rotor fluxes of an inverter-fed induction motor. Subsequently, the estimated stator currents are fed into an adaptive controller to track the desired stator current trajectories. The adaptive controller is constructed as a feedback signal (a nonlinear control law), depending on estimated stator currents supplied by the neural estimators and the desired reference trajectories to be tracked by the output. The control of the direct and quadrature components of the stator current successfully tracked a wide variety of reference trajectories after relatively short online training periods.

Online training of parallel neural network estimators for control of induction motors

A laboratory implementation of a fuzzy logic-tracking controller using a low-cost Motorola MC68HC11E9 microprocessor is described in this paper. The objective is to design the most optimal yet practical controller that can be implemented and marketed and which gives respectable performance, even when the system loads, inertia, and parameters are varying. A distinguishing feature of this work is the by-product goal of developing a marketable, simple, functional, and low-cost controller. The proposed fuzzy-logic controller (FLC) is inherently robust and is based on simple rules that are derived from engineering and experimental results. Several test cases have been conducted to confirm the effectiveness of the proposed controller. The results indicate excellent tracking performance for both speed and position trajectories. For the purpose of comparison, a bang-bang controller has been tested. The FLC performs significantly better than the traditional bang-bang controller. The bang-bang controller has been shown to be relatively inaccurate and lacking in robustness. A description of the implementation hardware system is also given.

M.D. Kankam

Papers

Development and implementation of an adaptive fuzzy-neural-network controller for brushless drives

A continually online-trained neural network controller for brushless DC motor drives

Experimental verification of a hybrid fuzzy control strategy for a high-performance brushless DC drive system

Online training of parallel neural network estimators for control of induction motors

Laboratory implementation of a microprocessor-based fuzzy logic tracking controller for motion controls and drives