IRANIAN JOURNAL OF ELECTRICAL AND COMPUTER ENGINEERING (IJECE)
Year:2007 | Volume:6 | Issue:2|Papers Count:12

This paper presents an application of self tuning fuzzy control technique for a Switched Reluctance Motor (SRM) Drive. In this research, the output-scaling factor (a) is adjusted in online by the gain tuning Fuzzy Logic Controller (GTFLC) rules according to the current status of SRM. The rule base of GTFLC for tuning the output SF is defined on error (e) and change of error (De) of the controlled variable using triangular equally overlapped membership functions (MF's). The proposed self-tuning technique is applied to PI- type FLC to conduct real time implementation of speed control for SRM. The proposed scheme shows a remarkably improved performance over its conventional PI type FLC. A prototype based on a 6/4 pole SRM drive has been constructed and used to practically verify the features of the proposed control strategy. The complete speed control algorithm of the SRM drive incorporating the STPIFLC is experimentally implemented and validated using a high-speed digital signal processor board TMS320F2812. It is shown that the presented STPIFLC for SRM drive is with fast tracking capability, less steady state error, robust to load disturbance and very less torque ripple.

In this paper a robust nonlinear controller is presented for Doubly-Fed Induction Machine (DFIM) drives. The nonlinear controller is designed based on combination of Sliding-Mode (SM) and Adaptive-Backstepping control techniques. Using the fifth order model of DFIM in a stator d, q axis reference frames with stator currents and rotor flux components as state variables, First a SM controller is designed in order to follow a linear reference model. Then, the SM control and adaptive backstepping control approach are combined to derive a robust composite nonlinear controller for DFIM, which makes the drive system robust and stable against the parameters uncertainties and external load torque disturbance. In this drive system two back-to-back voltage source two level SVM-PWM inverters are employed in the rotor circuit, to make the drive system capable of operating in motoring and generating modes below and above synchronous speed. Computer simulation results obtained, confirm the effectiveness and validity of the proposed control approach.

In this paper, an efficient and reliable Particle Swarm based approach to Optimal Design of three-phase Induction Motor (ODIM) is proposed. The proposed approach employs a particle-swarm-optimization (PSO) technique to search for optimal values induction motor design parameters. The proposed approach has been examined and tested on two sample motors with different objectives that reflect the motor's efficiency, active material cost, and performance under starting and full load conditions. The proposed approach results have been compared to those reported in the literature. The results are promising and show the effectiveness and robustness of the proposed approach.

The modern power system has become very complex in nature with conflicting requirements. Heavy load demands and distributed generation compromise the cost of generation. Also the pollution hazards and inherent losses contribute towards an inefficient system. A scenario of coordination between cost, emission and loss will echo a pareto-optimal generation. The oft handled iterative approaches and optimization techniques have been replaced by a Dynamic Programming technique involving novel recursive approach for realizing production cost minimization, with an emission constrained and loss reduced condition. A study of test system exhibits the computational efficiency and accuracy of the solution.

In traditional Adjustable Speed Drives (ASDs) voltage controller low power factor and high-energy are some of the major drawbacks. These can be eliminated using Genetic Algorithm based PWM optimization. This Genetic Algorithm optimization eliminates the harmonics in the inverter output and provides power factor close to unity. The increased use of power electronics-based loads, (adjustable speed drives, switch mode power supplies (SMPS), etc.) for improving system efficiency and controllability results in increase in the harmonic distortion levels in end use facilities and overall power system. In this paper, the harmonic contents and distortion factor are less in the output of Genetic Algorithm optimized pulse width Modulated impedance source inverter when compared to standard PWM and random PWM based inverters. It is the most economic methodology to improve the power quality of inverters.

The good knowledge and information about the fault mode behavior of the PWM voltage source inverter is important to improve system design, protection and fault tolerant control. This paper develops a real-time condition-monitoring algorithm for 3-phase pulse width modulation (PWM) inverter. It is designed to detect and identify the transistor open-circuit fault and simultaneous two transistors open-circuit fault, which commonly occurs in the inverter drive system. The condition monitoring mechanism is based on phases currents mean, maximum and minimum values as fault indicators. The developed system has been rigorously assessed theoretically and it has been shown that the system is robust and reliable. Simulation results are presented.

In this work some results related to the effect of insulating barriers on corona discharge are reported. It was clearly established that both the amplitude and the shape of the currents are strongly affected by the presence of organic barriers. The insulating barriers used in this work were constituted by layers of oil or pressboard paper with different thickness. The experiments were carried out in negative point-to-plane corona. The aim of the dc voltage, used in this work, was an examination of the I-V characteristics in order to determine the modifications in the discharge which can be related to electrochemical interaction involving activated species.

This paper presents a new coding technique for fingerprint compression, which is based on contourlet transform and Self-Organizing Feature Map (SOFM) vector quantization. One of the main difficulties in developing compression algorithms for fingerprints resides in the need for preserving the minutiae, which are subsequently used in identification. Wavelets have shown their ability in representing natural images that contain smooth areas separated with edges. However, wavelets cannot efficiently represent the ridge and furrow patterns, which are predominant in fingerprints. This issue is addressed by directional transforms, known as contourlets, which have the property of preserving edges. SOFM based vector quantizer quantizes the coefficients obtained by contourlet transform. The results obtained are tabulated and compared with those of the wavelet based ones.

A method for measuring 3D shape and motion parameters based on two approximately orthogonal X-ray angiographic image sequences synchronized with superposed electrocardiogram (ECG) signals is presented. The 3D vessel axis at each time-point is firstly recovered from the projective image pair. Clinically important configuration parameters, including diameter and length, are then measured based on 3D axes and 2D detection. The displacement field between 3D axes at consecutive time-points is estimated through minimizing a cost function in point-by-point manner. The performance of the approach has been validated with phantom data, computer-simulated data and clinical data. For diameters of the phantom ranging from 3 mm to 5 mm, errors are lower than 0.05 mm. For lengths ranging from 15 mm to 90 mm, errors are no more than 2.5 mm. Root mean square (RMS) displacement amplitude errors are all smaller than 0.01 mm for different transformations in validations with simulated data. Illustrative results clinically validated in real X-ray images indicate its feasibility and reliability for quantitatively measuring and interpreting arterial vessels.

Differential quadrature approximation and the Chebyshev collocation method are adopted to convert the partial differential equation that governs heat transfer into a lumped parameter model. The transient responses of a thermostat are calculated using the Runge-Kutta method. Differential quadrature approximation and the Chebyshev collocation method efficiently determinate the heat transfer properties. A control strategy for a fuzzy logic controller in the system is proposed. The fuzzy logic controller is implemented using assembly language and a single chip 89C51.

In this paper, we try to explain our approach to the problem of decision making in a multi-agent, non-deterministic environment, like soccer field. Our strategies have been tested in Robocup competitions, and could register important accomplishments. After reducing the problem in terms of the well-known problem of finding the best coefficients for computing the correctness of an action, we've used three different methods for calculation of the overall fitness, with respect environmental parameters. These methods are: getting simple weighted average with fixed weights, using GA to find the weights of the parameters, and finally using OWA operator instead of weighted averaging with GA. Due to the huge size of the problem we had to use techniques like bounded rationality and memorized estimation to reduce the amount of computation.

This paper studies the effect of nitrogen on a laser forming structure with strained quantum well containing GaInAs. In particular, it treats the question of evaluating the influence of nitrogen on the conduction band by using the anticrossing band model. The incorporation of nitrogen decreases the energy of the band gap and consequently increases the emission wavelength. The reduction in energy is due to the interaction between the conduction band energy and the nitrogen level, and the more the nitrogen concentration grows the more the band gap energy decreases. On the other hand the wavelength of emission also increases, thus the advantage of the incorporation of nitrogen in the proposed structure is to vary the emission wavelength in order to reach 1.3 mm. One can also exploit the effect of gallium composition and the width quantum well on the emission wavelength.