فصلنامه روشهای هوشمند در صنعت برق
پیاپی 5 (بهار 1390)

Locally Linear Model Tree (LOLIMOT) algorithm proposed by Nelles deals with local linear nearo-fuzzy models that is based on divides-and-conquer strategy that a complex modeling problem is divided to a number of smaller and thus simpler sub problems. So the characteristic of such a neuro-fuzzy model depends on division strategy for the original complex problem. For finding the best output the algorithm divides the problem to a number of local linear models (LLMs), then continues with finding the worst LLM and dividing it. LOLIMOT splits the local linear models into two equal halves with an axis-orthogonal decomposition strategy. In this paper a new approach based on extremeal optimization (EO) is used to optimize the structure of LOLIMOT. Simulation results show the effectiveness of the enhanced LOLIMOT to have a higher precision with optimal number of neurons.

Biometrics based personal identification is regarded as an effective method for automatic identification، with a high confidence coefficient. A multi modal biometric system consolidates the evidence presented by multiple biometric sources and typically provides better recognition performance compared to systems based on a single biometric modality. So in this paper we use combination of Face، Palm print and Ear characteristic to individual’s authentication. In our approach، features extracted using HMAX model are translation and scale-invariant. Then we applied Support vector machine and K-nearest neighbor classifiers to distinguish the classes. In fusion stage we use matching-score level. Experimental results showed 96% accuracy rate on ORL Face database and 96/6% accuracy rate on POLYU Palm print database and 97% accuracy rate on USTB Ear database; however we achieve 100% accuracy rate on Face، Palm print and Ear multi modal biometric.

In this paper we study a technique of power RF amplifier linearization predistortion digital circuits and propose a method for adapting this circuit and compare the advantages and disadvantages of this circuit with conventional predistortion circuit. The main problem with predisrortion circuit is the memory effect. We purpose a new circuit to overcome disrortion and memory effect problems based on DSP. Finally we compare results of our simulation with previous results. Our simulation shows a significant reduction in power ratio of adjacent channels.

Direct torque control with Current Source Inverter (CSI) instead of voltage source inverter is so appropriate because of determining the torque of induction motor with machine current and air gap flux. In addition، Space-Vector Modulation (SVM) is a more proper method for CSI because of low order harmonics reduction، lower switching frequency and easier implementation. This paper introduces the SVM method for CSI and uses the proposed inverter for vector control of an induction motor. The simulation results illustrate fast dynamic response and desirable torque and speed output. Fast and accurate response to changes of speed and load torque reference completely proves the prominence of this method.

Relevance feedback approaches is used to improve the performance of content-based image retrieval systems. In this paper، a novel relevance feedback approach based on similarity measure modification in an X-ray image retrieval system based on fuzzy representation using fuzzy attributed relational graph (FARG) is presented. In this approach، optimum weight of each feature in feature vector is calculated using similarity rate between query image and relevant and irrelevant images in user feedback. The calculated weight is used to tune fuzzy graph matching algorithm as a modifier parameter in similarity measure. The standard deviation of the retrieved image features is applied to calculate the optimum weight. The proposed image retrieval system uses a FARG for representation of images، a fuzzy matching graph algorithm as similarity measure and a semantic classifier based on merging scheme for determination of the search space in image database. To evaluate relevance feedback approach in the proposed system، a standard X-ray image database consisting of 10000 images in 57 classes is used. The improvement of the evaluation parameters shows proficiency and efficiency of the proposed system.

Nowadays، Kalman filter has been wildly used for solving the problem of real world. Kalman filter is a recursive filter that estimates the state of a linear dynamic system from a series of noisy measurements. One of the applications of Kalman filter is signal processing. In this paper، we use Kalman filter for electrocardiogram (ECG) signal noise removal. First accidental ECG signals are collected from Physiobank database and then Kalman filter is tuned for noise removing from ECG signals. In addition، we apply Finite Impulse Response (FIR) filter for ECG signal noise removing and finally we compare the performance of two filters using Mean Square Error (MSE) measurement. Results show the superior performance of Kalman filter for ECG signal noise removal.

One of the important issues in electric arc furnace (EAF) of the steel industries is the effective compensation of the EAF. In this paper، firstly an optimal design model for the EAF is presented. Then، a thyristor controlled reactor (TCR) is installed in parallel with the TCR. The parameters of the TCR are designed optimally to compensate the reactive power of the EAF. In the proposed method، a hyperbolic-exponential model is considered for the EAF and also a sinusoidal voltage flicker with the frequency in the range human vision (between 4-14 Hz) is included. In addition، the presence of the harmonic filters with the TCR is discussed. The results of the simulation show that the TCR can compensate the effects sinusoidal voltage flicker. Moreover، the results indicate that compared to the other methods، the average time response is about the half period of a cycle.

This paper describes an efficiency optimization control method for high performance interior permanent magnet synchronous motor drives with online estimation of motor parameters. The control system is based on an input-output feedback linearization method which provides high performance control and simultaneously ensures the minimization of the motor losses. The controllable electrical loss can be minimized by the optimal control of the armature current vector. It is shown that parameter variations except at near the nominal conditions have undesirable effect on the controller performance. Therefore، a parameter estimation method based on the second method of Lyapunov is presented which guarantees the stability and convergence of the estimation. The extensive simulation results show the feasibility of the proposed controller and observer and their desirable performances.