Motion control - Fault diagnosis in Machines using VHDL

The machines which have become a part of present day life, even in some cases it overcome human working ways. So for there is majo r concern in functioning of motor drives because a minute faulty function of motor may lead to drastic damage in working environment. Thus before entering into the fault diagnosis of induction motor the speed control is an important aspect. The speed control in induction motor can be done using SVPWM technique. From the past several years, much progress has been made in Art ificial Intelligence (AI) technology. Simp lified models of neural processing in the brain have been viewed as artificial intelligence in neural networks. It's an inexpensive, reliab le and non- invasive Artificia l Neural Network (ANN) based fault diagnosis. Multilayer Perceptron (M LP) is to be used in this paper because the input data contain continuous feature. The fault diagnosis in induction motor using AI technology can be done without resuming the function of induction motor an advantage of this approach compared with other techniques for fault diagnosis(1). This paper presents the new technique relevant to the design method of artificial intelligence based on VHDL hardware description language and FPGA imp lementation. The simu lation results are obtained from XILINX 12.2 software. I. Introduction In the past few years most of the methods are based on knowledge of status equation for fully or partially controlled systems. However status equation can't be easily obtained. Therefore we go for a smart control method with self-learn ing capability for better control performance. Thus here in this paper induction motor fault diagnosis and its motion control are going to be delt. Our design users Neural Netwo rk for its amazing effect which tradit ional controllers cannot achieve, when the system involved in an uncertain, time vary ing or non- linear status. The speed control of induction motor is the main process to be undergone before fault diagnosis is performed. The speed of the motor is of major concern in detection of fault. The speed control and fault diagnosis are performed in comparat ive manner (1). The speed control of induction motor is performed by space vector pulse width modulation (SVPWM). Thus three level voltage fed PWM inverter, which shows popularity in industrial drive applicat ion. The output voltage waveform which is generated by mult ilevel inverter. It can be generated at low switching frequency with high efficiency and low distortion. This is the reason why large voltage between the series device is easily shared. The space vector PWM technique is very popular which results in higher magnitude of fundamental output voltage available. The SVPWM is an advanced computation- intensive PWM method and it's the best among all other PWM method as it functions at variable frequency drive applications (2). II Speed Control of motor A voltage source inverter type Space vector pulse width modulation for controlling the speed of the induction motor is performed. The pulse width modulation in which several pulse are produced in each half cycle but width of each pulse is not same as in case of mult iple pulse width modulation however the width of each puls e is varied in accordance with the amplitude of the sine wave reference voltage. The Space Vector Modulation (SVM) is normally implemented using the direct method. The space Vector is an algorith m for the control of PWM. The principle behind SVPWM is that voltage vector which is to be approximated by using three adjacent vectors. To drive a 3phase AC powered motor at varying speed we require alternate current waveform wh ich can be generated using SVM. Thus Fig 1 shows the SVPWM wave model in response to the sinusoidal input wave. The SVPWM is generated mainly based upon the positive and negative cycles of the sinusoidal wave. The space vector diagram which consist 8 vectors these vectors are given as input to the IGBT inverter through which the speed of the induction motor is controlled. Taking into consideration of these 8 vectors among which 6 vectors starting from V001 to V110 is for supplying signal to Switch ON the transistors present in the IGBT inverter.

[1]  A. Kocalmis,et al.  Simulation of a Space Vector PWM Controller For a Three-Level Voltage-Fed Inverter Motor Drive , 2006, IECON 2006 - 32nd Annual Conference on IEEE Industrial Electronics.

[2]  Anton Kummert,et al.  FPGA implementation of a neural network for a real-time hand tracking system , 2002, Proceedings First IEEE International Workshop on Electronic Design, Test and Applications '2002.

[3]  H. Hikawa,et al.  Implementation of simplified multilayer neural networks with on-chip learning , 1995, Proceedings of ICNN'95 - International Conference on Neural Networks.

[4]  S. V. Dudul,et al.  Induction Machine Fault Detection Using Generalized Feed Forward Neural Network , 2009 .

[5]  P. Agarwal,et al.  Field Programmable Gate Array (FPGA) based neural network implementation of Motion Control and fault diagnosis of induction motor drive , 2008, 2008 IEEE International Conference on Industrial Technology.

[6]  Kamal Zuhairi Zamli,et al.  Hybrid version of MLP neural network for transformer fault diagnosis system , 2008, 2008 International Symposium on Information Technology.