University of Tabriz

About: University of Tabriz is a education organization based out in Tabriz, Iran. It is known for research contribution in the topics: Population & Nanocomposite. The organization has 12141 authors who have published 20976 publications receiving 313982 citations.

Boosting methods for multi-class imbalanced data classification: an experimental review

Abstract: Since canonical machine learning algorithms assume that the dataset has equal number of samples in each class, binary classification became a very challenging task to discriminate the minority class samples efficiently in imbalanced datasets. For this reason, researchers have been paid attention and have proposed many methods to deal with this problem, which can be broadly categorized into data level and algorithm level. Besides, multi-class imbalanced learning is much harder than binary one and is still an open problem. Boosting algorithms are a class of ensemble learning methods in machine learning that improves the performance of separate base learners by combining them into a composite whole. This paper’s aim is to review the most significant published boosting techniques on multi-class imbalanced datasets. A thorough empirical comparison is conducted to analyze the performance of binary and multi-class boosting algorithms on various multi-class imbalanced datasets. In addition, based on the obtained results for performance evaluation metrics and a recently proposed criteria for comparing metrics, the selected metrics are compared to determine a suitable performance metric for multi-class imbalanced datasets. The experimental studies show that the CatBoost and LogitBoost algorithms are superior to other boosting algorithms on multi-class imbalanced conventional and big datasets, respectively. Furthermore, the MMCC is a better evaluation metric than the MAUC and G-mean in multi-class imbalanced data domains.

A comprehensive review of low voltage ride through of doubly fed induction wind generators

Abstract: Wind power has become an important source of renewable energy in a number of countries around the world, including Denmark, Germany and Spain. Thereupon, connection of wind farms to the grid and their dynamic behavior under different grid conditions has become an important issue in recent years and new grid codes have been introduced. One of the most important issues related to grid codes is the low voltage ride through (LVRT) or fault ride through (FRT) capability of wind farms. Based on such code requirements, wind turbine generators must remain connected to the grid and actively contribute to the system stability during various grid fault scenarios that result in a generator terminal voltage dip. Moreover, wind turbine generators should have the ability to supply reactive power during the faults. In addition, they should supply active and reactive power immediately after fault clearance to support the network frequency and voltage, respectively. In this paper, a comprehensive review of researches published about analysis, modeling and improvement of LVRT of wind turbines with doubly fed induction generator (DFIG) is presented. The review also concludes that more investigations should be carried out to completely fulfill the grid codes׳ requirements. In particular, reactive and active power requirements of grid codes should be taken into account in more depth in the future LVRT solutions.

Entropy Generation in a Circular Tube Heat Exchanger Using Nanofluids: Effects of Different Modeling Approaches

Abstract: The main goal of this paper is to compare single- and two-phase modeling approaches for forced convection flow of water/TiO2 nanofluid. The considered geometry is a horizontal tube with constant wall heat flux boundary condition where flow regime is turbulent. A computational fluid dynamics (CFD) approach is utilized for heat transfer and flow field estimation of the single-phase and three different two-phase approaches, namely, volume of fluid, mixture, and Eulerian models. Results are presented for Reynolds numbers ranging from 9000 to 21,000, for different nanoparticle diameters ranging from 20 to 40 nm, and for values of volume fractions ranging from 0 to 4%. The obtained results show that the values of entropy generation for thermal and turbulent dissipation are very close for the single-phase and mixture models. Numerical investigation showed that the values of entropy production for pure water are identical regardless of the CFD approach; however, when the volume fraction of nanoparticles i...