Amirkabir International Journal of Electrical & Electronics Engineering
Volume:53 Issue: 1, Winter-Spring 2021

In this paper, an optimal adaptive robust pitch controller is proposed for variable speed wind turbines (VSWTs). The proposed pitch controller has stability analysis, meanwhile it simultaneously keeps the generated power of the wind turbine at the rated power and mitigates the mechanical loads on the gearbox. The proposed pitch controller in this paper has two terms. The first term is a radial basis function neural network (RBFNN), to approximate unknown nonlinear functions of the wind turbine. Another term is a chattering-free continuous robust structure, which can cope with the approximation error. The weights of RBFNN and the gain of the robust structure are derived via the Lyapunov synthesis approach. It is proved that the closed-loop signals are semi-globally uniformly ultimately bounded. The optimal parameters of the proposed controller are derived by solving a proposed multi-objective optimization problem using non-dominated sorting genetic algorithm-II (NSGA-II) and multi-objective particle swarm optimization (MOPSO) algorithm. The effectiveness of the proposed controller is compared with the baseline PI controller designed by NREL. First, both the proposed and baseline PI controllers are applied to the general model (2-mass model) of the wind turbine and then they are validated via a highly reliable simulator called FAST. The results demonstrate the effectiveness and applicability of the proposed pitch controller.

In this paper, two different methods are introduced for human activity recognition based on video signals. Method 1 explores the effectiveness of combining feature descriptors obtained by local descriptors and artificial neural network classifier. It stays in traditional approach that is local descriptors extract interest points or local patches from videos, then feature vectors are constructed based on them, and eventually feature vectors are used as the input of a two-layer feed-forward artificial neural network (ANN). Experimental results show that using HOG3D descriptor with ANN gives the best performance. On the other hand, deep learning architectures have attracted much consideration in the last years for automatic feature extraction, so an improved 3D convolutional neural network architecture is also designed as method 2. They are implemented and compared with state-of-the-art approaches on two data sets. The results exhibit that method 1 is superior when the shortage of sample data is the main restriction. It achieves recognition accuracies of 97.8% and 99.8% for the Weizmann and KTH action data sets, respectively. Also method 2 is considerable because of its automatic features extraction and achieves an acceptable result for video with lots of original training data. So that it gets recognition accuracy of 92% for the KTH data set while this value is drastically reduced for the Weizmann data set.

Power system flexibility is the ability of power system to cope with the uncertainty and variability both in generation and load sides. This ability should be quantified and measured by a suitable index to show the level of system flexibility in different situations. Flexibility area index, proposed by the authors is a suitable metric for power system flexibility evaluation especially in the presence of renewable sources as large scale wind and solar farms. Similar other system flexibility indices, this index is defined at first for one generation unit and then extended to the power system by combination the unit indices. In this way an accurate and meaningful combination routine should be established to reflect the effect of each unit flexibility index in the combined system flexibility index correctly.This paper proposes a suitable and justified method to combine the unit flexibility indices achieving the system flexibility index. The performance of the proposed index is verified by the wind/load curtailment in economic load dispatch incorporated wind power. Achieving this purpose, the mentioned index is decomposed into two components, one for ramp up and maximum generation system capabilities (upper component) and another for ramp down and minimum generation system capabilities (lower component) each of them related to the load or wind curtailment respectively which is another contribution of this paper. Finally by establishment a correlation between upper/lower component and load/wind curtailment, a suitable validity evaluation for the proposed system flexibility index is done which is another contribution of this paper.

Most of the studies on phenotype differences, including some diseases, are based on studying the areas of the genome called Single Nucleotide Polymorphism (SNP). Some SNPs on their own and some by interacting with other SNPs play an important role in any phenotype or specific disease. Various models, including regression models, are designed and implemented for prediction of these diseases. As the phenotypes are both quantitative and binary, linear regression is used for models predicting quantitative ones, which is only based on the number of minor alleles per SNP, and logistic regression is used for binary ones like complex diseases. Since complex diseases are not caused only by independent SNPs, but by the interaction of a large number of SNPs, which mostly exceeds the number of samples, penalized logistic regressions are counted to be a better choice. These models, therefore, can overcome the limitation of ordinary logistic regression on high-dimensional SNP datasets. In this paper, three regression models, including Ridge, Lasso and Elastic Net (EN), were implemented on 10000 samples of the SNP datasets of OWKIN-Inserm Institute to predict the risk of a specific disease (undisclosed for confidentiality reasons). Among these three, the Lasso model with minimizer lambda indicated higher accuracy (73.73%) and AUC (83.54%). The model is also less complex since it eliminates less related features as much as possible and keeps only the most informative ones. Besides, getting better results with Lasso indicates that multicollinearity is either not existed between variables or is low that can be neglected.

The densely packed decimal (DPD) encoding for secondary and primary storage of three binary coded decimal (BCD) digits is included in the IEEE 754-2019 standard for decimal floating-point arithmetic. Binary coded chiliad (BCC) representation of 3 BCD digits (i.e. radix-1000) will achieve equi-efficient packing. The primary advantage is that BCC operands can be directly manipulated by arithmetic operations, while DPD operands have to undergo DPD-to-BCD and reverse conversions afore and ahead of each arithmetic operation. We are therefore interested in designing the arithmetic unit that receives BCC operands and produces BCC results, following previous BCC works. Compared to the equivalent BCD or other radix-10 arithmetic, prospects show that at least equally efficient arithmetic units are feasible for BCC arithmetic, as even better performance has been achieved in the case of addition. Therefore, we demonstrate the IEEE 754-2019 compatibility of the BCC Encoding in this paper. Consequently, for the DPD-to-BCD expansion and the reverse compression, the DPD-to-BCC converter, and the reverse blocks, we show the delay, area, and power dissipation. The findings show a substantial delay (83%), area (27%), and power (29%) overhead. However, as the number of conversions in the latter case is much less than that of the former, overall power dissipation is expected to decrease considerably.

Fault-Tolerant Hybrid Excited Axial Field Flux-Switching (FT-HEAFFS) motor is a new type of doubly salient stator-type permanent magnet motor, which combines the advantages flux switching motor and hybrid excitation motor. This motor has compact structure, high power density, high efficiency and strong anti-demagnetization capability. The additional excitation winding makes the air gap magnetic field adjustable, which can increase the output torque and extend the speed range. It is suitable for use in the system of frequency conversion and speed regulation of electric vehicles. To improve the performance of the fault-tolerant-hybrid excitation axial field flux-switching (FT-HEAFFS) motor and attain the minimum copper loss, a fault-tolerant control method based on model predictive control algorithm is proposed. Considering a 6 stator slots/13- rotor poles FT-HEAFFS machine as the control object, under the open circuit failure of single-phase winding, the minimum cropper loss fault-tolerant method based on the model predictive torque control (MPTC) and direct torque control are studied and analysed, respectively. The feasibility and effectiveness of the proposed fault-tolerant control method are verified. The research results showed that both methods could make the speed, torque and stator flux-linkage almost unchanged, ensuring the stable operation of the system. Compared with direct torque control, the model predictive flux control had smaller flux-linkage ripple before and after the open circuit failure.

Despite the development of the use of frequency response analysis (FRA) in condition monitoring of power transformers, how to interpret the results of FRA measurements has not yet been standardized. Therefore, proposing new methods to interpret the results of FRA measurements in research works is followed with great interest by researchers. This paper proposes a k-nearest neighbor (k-NN) based method for condition monitoring of transformers using the results of FRA measurements. First, the necessary measurements are performed on healthy and faulty transformers (under different fault conditions) and the required database is created. Then, by extracting the peak (resonance) and trough (anti-resonance) points of the measured transfer functions from the transformer, several mathematical features for training and validation of k-NN are extracted. Finally, by applying the data obtained from actual transformers, the performance of k-NN in different states is evaluated and compared. The results show that the proposed method is able to determine the condition of the transformer (whether it is healthy or defective) with very good accuracy and if it is defective, identify the type of defect. In addition, in order to prove the ability of k-NN, a comparison is made with the results of the artificial neural network (ANN).

In this paper, we introduce a new linear received signal strength-based estimator for unknown node localization which its accuracy at low Signal-to-noise ratio (SNR) is better than many linear estimators and can compete with estimators based on the convex optimization, but it is much lighter than convex optimization-based estimators. The main ingredients in our proposed linear position estimator are to reformulate the localization problem in terms of Tikhonov-regularization and introduce a biased noise variable. The way that we apply for this reformulation avoids any possible linear approximation in which target position variables are involved, thus saving fair amount of information. The way we use to approximate the nonlinear terms leads us to a linear estimator to outperform most convex estimators in terms of both accuracy and speed in high power noise. The proposed algorithm is also indifferent to the transmit power and thus applicable to either known or unknown transmit power scenarios. In other words, another advantage of our scheme is its ability to compute a closed-form expression for estimate of as a by-product.Simulation results show the efficacy of the proposed algorithm in comparison to other methods for both typical RSS-based measurement data model and the modified model for indoor application.

Considering the air/earth interface to compute the electromagnetic field at the desired point is the most important problem in transient analysis of grounding system buried to lossy media like earth. In order to consider this issue in the available proposed method and to obtain a solution for the problem, one needs to account an integral solution of the so-called Semmerfeld integral in the cylindrical coordinate system. Analytical solution for such integral almost impossible which is due to the presence of the oscillating the zeroth-order Bessel function of the first kind and also singularities and branch-cuts in its integrand function. In this paper, we investigate the behavior of the vertical grounding electrode in a high-frequency electromagnetic transient state based on the near field theory using the method of moments (MoM). To compute the electromagnetic field at the desired point, the main problem is calculating the well-known Sommerfeld integral in the cylindrical coordinate system which its integral kernel includes the zeroth-order Bessel function of the first kind along with some singularities and branch-cuts. Since the analytical solution of this integral is not available in literature, we propose a numerical method, as well as a strategy based on the exact solution for far and near filed calculations. A detailed analysis of the obtained results compared with other techniques are provided to confirm the accuracy and validity of the proposed method.

Different surface impedance models are applied to circular nano-wires at terahertz and optical frequencies and the accuracy of these surface impedance boundary conditions (SIBCs) is studied. The simplest form of SIBC defines a local relation between the tangential electric and magnetic equivalent surface currents at each point on the boundary. This definition is very dependent on the constituent material of the wire and its radius. The generalized IBC (GIBC) improves the accuracy of the local definition by considering the curvature of the surface at each observation point. On the other hand, the operator definition of surface impedance presented in the SIGO method (surface impedance generating operator), is an exact field theoretical approach that determines the relation between equivalent electric and magnetic surface currents. Moreover, this method is suitable for parallel processing. For the special case of circular wires, the SIGO operator is derived. To validate the SIBC models, the results are compared with the SIGO. In spite of its extreme simplicity, it is observed that the accuracy of SIBC models is limited at optical and terahertz frequencies. It is also shown that some forms of SIBCs presented in the literature for nano-wires can be considered as special cases of SIGO formulation.

Distribution Network Reconfiguration (DNR) is an important challenge in the operation of distribution networks which may be influenced by factors such as Wind Turbine Generators (WTG). In this paper, a novel policy is implemented to solve the DNR problem in presence of WTGs. The objectives of proposed DNR policy are minimization of active power losses, total electrical energy costs, and total emissions of the network. To solve the problem, an improved version of Honey Bee Mating Optimization (IHBMO) algorithm is implemented. Moreover, a stochastic scenario-based model is considered to meet the uncertainty of WTGs and loads. The bases of proposed stochastic model are generation of stochastic scenarios using the roulette wheel mechanism, and a scenario reduction technique to decrease the computation burden of the problem. For each scenario, a multi-objective mechanism is employed to save non-dominated solutions extracted by IHBMO. A decision-making procedure based on fuzzy clustering technique is used to rank the obtained non-dominated solutions according to the decision-maker preferences. Finally, an 84-bus distribution test network is considered to evaluate the feasibility and effectiveness of the proposed method. Obtained results show that proposed method can be a very promising potential method for solving the stochastic multi-objective reconfiguration problem in distribution systems.