@article { author = {}, title = {Robust sensorless vector control of induction machines}, journal = {Iranian Journal of Science and Technology Transactions of Electrical Engineering}, volume = {33}, number = {2}, pages = {133-147}, year = {2009}, publisher = {Shiraz University}, issn = {2228-6179}, eissn = {}, doi = {10.22099/ijste.2009.832}, abstract = {A sensorless vector control strategy for induction machine (IM) operating in variable speed systems is presented. The sensorless control is based on a reduced-order linear observer based on terminal voltage and current as input signals. An estimation algorithm based on this observer is proposed to compute speed. It is shown that the proposed sensorless control is more sensitive to the stator resistance than to the rotor resistance. In order to tune the observer and to compensate for the parameter variations and the uncertainties, a separate estimation of the stator resistance is introduced. The equations to estimate the stator resistance are derived from the machine differential equations. For certain operating regions of the machine, it is verified that the stator resistance can be accurately estimated regardless of wide stator resistance variation. It is shown that design and hardware implementation of this method is simpler than the previous works. The simulation and experimental results demonstrate the good performance of the proposed observer and estimation algorithm and of the overall indirect-field-oriented-controlled system.           }, keywords = {Sensorless control,Induction machine,observer,parameter identification}, url = {https://ijste.shirazu.ac.ir/article_832.html}, eprint = {https://ijste.shirazu.ac.ir/article_832_a0a2d39b5b0620052d38d2b7d3187768.pdf} } @article { author = {}, title = {Rotor position detection with employment of the pulse injection technique on switch capacitors in switched reluctance motors}, journal = {Iranian Journal of Science and Technology Transactions of Electrical Engineering}, volume = {33}, number = {2}, pages = {149-161}, year = {2009}, publisher = {Shiraz University}, issn = {2228-6179}, eissn = {}, doi = {10.22099/ijste.2009.833}, abstract = {This paper presents a new and novel method designed to detect rotor position by decoding received information in two different techniques for switched reluctance motor at standstill and also low speeds. These decoding techniques are namely, digital pulse width modulation (DPWM) decoder and digital pulse position modulation (DPPM) decoder. This method begins by connecting a small capacitor in series with each motor phase winding. The supply voltage in conjunction with the drive transistor produces a short pulse for each phase capacitor. Since the phase inductance varies with the rotor position, the rise time of the voltage produced on each capacitor changes as well. Next, the capacitor voltages are compared with some pre-set voltage values and the rotor position is detected at standstill mode. The number of pre-set voltage values varies for each technique. It then continues to sense the rotor position while the motor is running by applying the same procedure, but only to the un-energized phases windings. The results obtained demonstrate the feasibility and practicability of the method}, keywords = {Switched reluctance motor,sensorless,pulse injection,DPWM,DPPM}, url = {https://ijste.shirazu.ac.ir/article_833.html}, eprint = {} } @article { author = {}, title = {A simplified droop method implementation in parallel UPS inverters with proportional-resonant controller}, journal = {Iranian Journal of Science and Technology Transactions of Electrical Engineering}, volume = {33}, number = {2}, pages = {163-178}, year = {2009}, publisher = {Shiraz University}, issn = {2228-6179}, eissn = {}, doi = {10.22099/ijste.2009.834}, abstract = {In this paper, a simpler implementation of the well-known droop method for the control of parallel Uninterruptible Power Supply (UPS) systems is presented. In this method, in the power-sharing control scheme, the output current is calculated by software without the need for a current sensor, resulting in a simpler and cheaper structure. By doing so, the number of feedback sensors is reduced from three to two. The paralleling strategy uses the droop method in which the control strategy is based on the drop in the inverter output frequency and amplitude. The application of Proportional-Resonant (PR) controllers is also extended to parallel inverter and its superior performance over the well-known Proportional-Integral-Derivative (PID) controller is shown. To show the performance of the proposed system, a system of two-parallel connected UPS is simulated, and two types of linear and non-linear loads are considered. The non-linear load is compliant with the IEC 62040-3 standard for class I UPS. The results show that the reduction of sensors results in no error, and the control system performance is quite satisfactory. To verify the proposed concept, a two-625VA UPS system is implemented. Several tests on both linear and non-linear loads are performed and the results, which are in good agreement with those of the simulations, are provided. The results indicate that the proposed parallel inverter control structure provides a better system in terms of performance parameters.           }, keywords = {Uninterruptible power supply (UPS),droop control method,proportional-resonant (PR) controller}, url = {https://ijste.shirazu.ac.ir/article_834.html}, eprint = {https://ijste.shirazu.ac.ir/article_834_30fb4817018fcdeb38ba5f0030487922.pdf} } @article { author = {}, title = {Surge suppression using a fuzzy hysteresis-band controlled resistor array}, journal = {Iranian Journal of Science and Technology Transactions of Electrical Engineering}, volume = {33}, number = {2}, pages = {179-189}, year = {2009}, publisher = {Shiraz University}, issn = {2228-6179}, eissn = {}, doi = {10.22099/ijste.2009.835}, abstract = {Transient over voltage protection has always been an important part of the electrical design of power systems.  Metal–oxide varistor (MOV) has been a useful tool to limit the impulse voltage stress in recent years. This device has some disadvantages: 1. Its protection voltage is selected a little higher than the system nominal voltage. The main reason is the variation of the power frequency peak voltage, 2. It is only a two-valued resistor (switches between high and low impedances), 3. MOV has a relatively large slope resistance in conduction mode, 4. MOV’s protection process depends only on the voltage amplitude and has no obvious dependence on the variation rate of the voltage amplitude, 5. Threshold voltage of a MOV is a constant value and depends on its physical parameters. This paper presents a method to alleviate the aforementioned disadvantages and introduces a resistor array controlled by a fuzzy hysteresis-band controller to achieve a variable resistor. The first disadvantage is addressed using a power frequency voltage amplitude detector. Appropriate values of resistors and a switching strategy to parallel suitable resistors in the resistor array can lead to diverse values of the needed resistor. In this method load voltage derivative is utilized to help surge suppression and the threshold voltage has an adaptive variance due to power frequency voltage amplitude. It is also shown that the proposed circuit has negligible leakage current.           }, keywords = {Surge suppressor,fuzzy control,hysteresis-band control}, url = {https://ijste.shirazu.ac.ir/article_835.html}, eprint = {https://ijste.shirazu.ac.ir/article_835_a8f75875bf9d557604cfb01b48728544.pdf} } @article { author = {}, title = {Automatic blood vessel segmentation in color images of retina}, journal = {Iranian Journal of Science and Technology Transactions of Electrical Engineering}, volume = {33}, number = {2}, pages = {191-206}, year = {2009}, publisher = {Shiraz University}, issn = {2228-6179}, eissn = {}, doi = {10.22099/ijste.2009.836}, abstract = {Automated image processing techniques have the ability to assist in the early detection of diabetic retinopathy disease which can be regarded as a manifestation of diabetes on the retina. Blood vessel segmentation is the basic foundation while developing retinal screening systems, since vessels serve as one of the main retinal landmark features. This paper proposes an automated method for identification of blood vessels in color images of the retina. For every image pixel, a feature vector is computed that utilizes properties of scale and orientation selective Gabor filters. The extracted features are then classified using generative Gaussian mixture model and discriminative support vector machines classifiers. Experimental results demonstrate that the area under the receiver operating characteristic (ROC) curve reached a value 0.974, which is highly comparable and, to some extent, higher than the previously reported ROCs that range from 0.787 to 0.961. Moreover, this method gives a sensitivity of 96.50% with a specificity of 97.10% for identification of blood vessels}, keywords = {Retinal blood vessels,Gabor filters,support vector machines,vessel segmentation}, url = {https://ijste.shirazu.ac.ir/article_836.html}, eprint = {https://ijste.shirazu.ac.ir/article_836_8b3f9ef0f94754664ab71a9b21455d12.pdf} }