Marcelo Suetake

Embedded DSP-Based Compact Fuzzy System and Its Application for Induction-Motor

This paper presents a compact embedded fuzzy system for three-phase induction-motor scalar speed control. The control strategy consists in keeping constant the voltage-frequency ratio of the induction-motor supply source. A fuzzy-control system is built on a digital signal processor, which uses speed error and speed-error variation to change both the fundamental voltage amplitude and frequency of a sinusoidal pulsewidth modulation inverter. An alternative optimized method for embedded fuzzy-system design is also proposed. The controller performance, in relation to reference and load-torque variations, is evaluated by experimental results. A comparative analysis with conventional proportional-integral controller is also achieved.

V/f

This work presents an alternative approach based on neural network method in order to estimate speed of induction motors, using the measurement of primary variables such as voltage and current. Induction motors are very common in many sectors of the industry and assume an important role in the national energy policy. The nowadays methodologies, which are used in diagnosis, condition monitoring and dimensioning of these motors, are based on measure of the speed variable. However, the direct measure of this variable compromises the system control and starting circuit of an electric machinery, reducing its robustness and increasing the implementation costs. Simulation results and experimental data are presented to validate the proposed approach.

Speed Control

This paper presents a methodology for the implementation of embedded fuzzy system algorithms to be built in digital signal processors. For the purpose of this application, the technique was applied for induction motor scalar speed control. The adopted control strategy is to adjust the amplitude and frequency (V/f) of fundamental supply voltage signal of induction motor, which is achieved by a three-phase pulse width modulation inverter. The fuzzy control system was therefore entirely embedded in DSP by applying simplification techniques, which aim at computational cost and memory requirements reduction. The controller performance in relation to load torque and speed reference variations was evaluated by experimental tests. A comparative analysis with conventional PI and PID controllers was also achieved.

Speed estimation for induction motor using neural networks method

Intelligent Systems are able technical of incorporate knowledge and, therefore, are being employed in different areas, improving and innovating conventional methods. As an example, the presence of artificial intelligence in monitoring systems to identify faults in electric motors. The purpose of such systems is to prevent unscheduled maintenance or avoid significant losses in the production line. Therefore, this paper describes the performance of two topologies of neural networks for identification of short circuit in the stator windings and bearing failures. The input data to the neural networks are statistical parameters extracted from on power supplies induction motor. Thus, the intelligent system proposed in this paper proved to be efficient and able to be implemented in monitoring systems failures in induction motors.

Sistema fuzzy compacto embarcado em DSP e sua aplicação para controle V/f de motores de indução

A neural method is presented in this paper to identify the harmonic components of an AC controller. The components are identified by analyzing the single-phase current waveform. The method effectiveness is verified by applying it to an active power filter (APF) model dedicated to the selective harmonic compensation. Simulation results using theoretical and experimental data are presented to validate the proposed approach.

Neural network classifier for faults detection in induction motors

Many researches have been developed in order to design intelligent systems for diagnosing faults in electrical machines. These faults involve since electrical problems, such as short circuit in one phase of the stator, until mechanical problems, such as bearing damage. Among the intelligent systems applied in these situations, we highlight the artificial neural networks, the fuzzy inference systems, the genetic algorithms and the hybrid systems, such as those neuro-fuzzy approaches. This paper describes an overview of the most relevant works in this area, which have obtained promising results from the use of intelligent systems in the processes involved to the fault diagnosis in electrical motors. Its main contribution is in providing several technical aspects in order to address future works in this application field.

A neural networks-based method for single-phase harmonic content identification

This work proposes an artificial neural network approach to estimate the induction motor speed. The induction motor has a great importance in many industrial sectors in which most control strategies are based on speed measurement of the motor axis. However, the direct measurement of this quantity reduces its robustness, compromises the driver system and control, besides increasing the implementation cost. The presented strategy estimates the induction motor speed when it is driven using scalar control. Simulation results are shown to verify the performance of the proposed method when the induction motor is submitted to load torque variations.

Sobre a aplicação de sistemas inteligentes para diagnóstico de falhas em máquinas de indução - uma visão geral

This paper presents a single-phase switched reluctance machine operating as a generator. For that purpose, a computational model was developed in Matlab Simulink environment, wherein all the system components, such as voltage source, drive and machine model, and load were integrated. The current and voltage behavior for several points of operations are presented. Furthermore, experimental tests were also carried out in a simple prototype to validate its functionality and simplicity of operation, providing a stable power generation over a wide range of speed. The results showed that the single-phase switched reluctance generator can be robust, efficient, and promising for especial applications.

A neural speed estimator in Three-Phase Induction Motors powered by a driver with scalar control

Many electronic drivers for the induction motor control are based on sensorless technologies. Load monitoring and failure prediction are also based in this methodology. The proposal of this work is to present a comparative study between single and multilayer neural networks used as an alternative approach of speed estimation, from transient to steady state. Simulation results are presented to validate the proposed approach.

Investigations on single-phase switched reluctance generator under variable speed operation

This paper proposes a fuzzy control system that allows simulating experimental industrial loads coupled to electric machines rotor shaft. The main focus of the application relies on the improvement of a test bench that enables experimental essays before inserting the studied electric machine in the industrial process. In this paper, a direct current generator is coupled to a three-phase induction motor rotor shaft. The imposed load torque is generated by controlling the field winding supply voltage using PWM techniques. Experimental results are presented in order to analyze and validate the proposed system performance. Comparative analysis with conventional PID controller is also considered in this paper.