Article
Computer Science, Theory & Methods
Akbar Telikani, Amirhessam Tahmassebi, Wolfgang Banzhaf, Amir H. Gandomi
Summary: Evolutionary Computation approaches, inspired by nature, provide a reliable and effective way to address complex problems in real-world applications. They have been used to improve machine learning models and quality of results, contributing to addressing challenges in the field.
ACM COMPUTING SURVEYS
(2021)
Article
Computer Science, Artificial Intelligence
Yuan Yuan, Wolfgang Banzhaf
Summary: We propose a new surrogate-assisted evolutionary algorithm for expensive multiobjective optimization. The algorithm uses two classification-based surrogate models, addresses dominance prediction problem using deep learning techniques, and integrates the surrogate models with multiobjective evolutionary optimization using a two-stage preselection strategy. Experimental results show the superiority of the proposed algorithm compared with several representative surrogate-assisted algorithms.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2022)
Article
Computer Science, Artificial Intelligence
Zhi-Hui Zhan, Jun Zhang, Ying Lin, Jian-Yu Li, Ting Huang, Xiao-Qi Guo, Feng-Feng Wei, Sam Kwong, Xin-Yi Zhang, Rui You
Summary: This paper proposes a matrix-based EC (MEC) framework for efficiently solving large-scale or super large-scale optimization problems. By defining the entire population as a matrix, the parallel computing functionalities of matrix can accelerate the computational speed of evolutionary operators. MEC is a promising way to extend EC to complex optimization problems.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2022)
Article
Automation & Control Systems
Qinqin Fan, Yilian Zhang, Ning Li
Summary: The paper introduces an automatic selection strategy of multiobjective evolutionary algorithms based on performance indicators (MOEAS-PI). This strategy can effectively improve the efficiency and robustness of solving multiobjective optimization problems.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2022)
Editorial Material
Computer Science, Artificial Intelligence
Uwe Aickelin, Hadi Akbarzadeh Khorshidi, Rong Qu, Hadi Charkhgard
Summary: This special issue focuses on the application of multiobjective evolutionary optimization in machine learning. Optimization plays a crucial role in many machine-learning techniques, and there is still potential to further utilize optimization in machine learning. Each machine-learning technique has hyperparameters that can be adjusted through evolutionary computation and optimization, considering multiple criteria such as bias, variance, complexity, and fairness in model selection. Multiobjective evolutionary optimization can help meet these criteria for optimizing machine-learning models. Although some existing approaches transform the problem into a single-objective optimization problem, multiobjective optimization models are more effective in contributing to multiple intended objectives or criteria.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2023)
Article
Automation & Control Systems
Huangke Chen, Ran Cheng, Witold Pedrycz, Yaochu Jin
Summary: This paper proposes a method to solve multiobjective optimization problems through multi-stage evolutionary search, highlighting convergence and diversity in different search stages. The algorithm balances and addresses the issues in multiobjective optimization through two stages.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2021)
Article
Automation & Control Systems
Ye Tian, Cheng He, Ran Cheng, Xingyi Zhang
Summary: This article provides a detailed explanation of existing diversity preservation approaches in MOEAs and their limitations, and proposes a multistage MOEA to address these limitations for better diversity performance.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2021)
Article
Mathematics
Omer Ali, Qamar Abbas, Khalid Mahmood, Ernesto Bautista Thompson, Jon Arambarri, Imran Ashraf
Summary: This study introduces a competitive coevolution process to enhance the capability of Phasor PSO (PPSO) for global optimization problems. Experimental results show that the improved competitive multi-swarm PPSO (ICPPSO) algorithm achieves a dominating performance, with average improvements of 15%, 20%, 30%, and 35% over PPSO and FMPSO.
Article
Computer Science, Artificial Intelligence
Xiangyu Wang, Bingran Zhang, Jian Wang, Kai Zhang, Yaochu Jin
Summary: This paper proposes a novel cluster-based competitive particle swarm optimizer equipped with a sparse truncation operator for solving sparse multi-objective optimization problems. Experimental results show that the proposed algorithm outperforms its peers on sparse test instances and neural network training tasks.
SWARM AND EVOLUTIONARY COMPUTATION
(2022)
Article
Automation & Control Systems
Izaz Ur Rahman, Zidong Wang, Weibo Liu, Baoliu Ye, Muhammad Zakarya, Xiaohui Liu
Summary: This study presents a novel N-state Markovian jumping PSO algorithm based on the evolution of states governed by a Markov chain, which outperforms some popular PSO algorithms in solving optimization problems.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Shuai Wang, Jing Liu, Yaochu Jin
Summary: The robustness of complex networks is crucial for their stability, and multiobjective robustness optimization is gaining increasing attention. However, challenges remain, including different computational complexities, insufficient network diversity, and high computational costs. By introducing a computationally efficient multiobjective optimization algorithm, a unique feature-based fitness evaluation method, and a surrogate ensemble based on graph embedding information, these challenges have been successfully addressed.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2021)
Article
Mathematics
Martin Montes Rivera, Carlos Guerrero-Mendez, Daniela Lopez-Betancur, Tonatiuh Saucedo-Anaya
Summary: Optimizing large-scale numerical problems is challenging, but our proposed improved PSO algorithm (DSRegPSO) has achieved significant success in reducing stagnation in local optimal regions.
Article
Computer Science, Artificial Intelligence
Maria-Luisa Perez-Delgado, Mehmet Akif Gunen
Summary: This article compares various evolutionary computation and swarm-based methods for solving the color quantization problem. Ten metaheuristics and four classical methods were compared using benchmark images and different palette sizes. The results show that swarm-based methods outperform classical methods in both qualitative and quantitative evaluations.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Engineering, Civil
Arya Yaghoubzadeh-Bavandpour, Omid Bozorg-Haddad, Mohammadreza Rajabi, Babak Zolghadr-Asli, Xuefeng Chu
Summary: Real-world problems are complex and classical optimization techniques may struggle to find optimal solutions. This study compares swarm intelligence (SI) and evolutionary computation (EC) algorithms, as well as nature-based and human-based algorithms, in the context of water resources planning and management. The results indicate that SI algorithms outperform EC algorithms in terms of solution accuracy, convergence rate, and run time.
WATER RESOURCES MANAGEMENT
(2022)
Article
Computer Science, Artificial Intelligence
Taiyong Li, Jiayi Shi, Wu Deng, Zhenda Hu
Summary: This paper proposes a novel particle swarm optimization algorithm called PPSO, which utilizes a pyramid structure and competitive-cooperative strategies to update particle information. Extensive experiments demonstrate that PPSO outperforms other algorithms in terms of accuracy and convergence speed, indicating its potential in numerical optimization.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Information Systems
Gilsu Choi, Gerd Bramerdorfer
Summary: This paper presents a detailed process for designing interior permanent magnet synchronous machines for electric vehicle applications. It proposes a comprehensive design process with two objectives and three constraints to achieve an optimal design that meets the key performance requirements for EVs. The impact of machine dimensions on key design parameters is thoroughly investigated, and additional analyses are provided to highlight the advantages and limitations of different designs.
Article
Energy & Fuels
Haipeng Liu, Xin Jin, Nicola Bianchi, Gerd Bramerdorfer, Pengzhong Hu, Chengning Zhang, Yongxi Yang
Summary: This paper proposes an approach to mitigate the cogging torque of permanent magnet machines by combining uncertainties in a specific sequence, achieving stable performance without significant cost increase.
Article
Engineering, Multidisciplinary
Ornella Stiscia, Sandro Rubino, Silvio Vaschetto, Andrea Cavagnino, Alberto Tenconi
Summary: This article proposes a methodology for computing efficiency maps of induction machines in wide torque-speed ranges. The modeling approach is based on the equivalent circuit of the machine, and the circuit parameters are determined through no-load and locked-rotor tests.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2022)
Article
Engineering, Multidisciplinary
Giacomo Scelba, Daniele Cremente, Giulio De Donato, Silvio Vaschetto, Emmanuel B. Agamloh, Andrea Cavagnino
Summary: Recent progress in wide band gap (WBG) semiconductor technology has made it feasible to have three phase inverters with switching frequencies above 100 kHz, driven by the desire to achieve extraordinary power densities. However, the impact of these inverters on the ac motors they supply has not been extensively discussed. This article investigates the effects of high switching frequency inverters on induction motors, highlighting the previously unreported impact of common mode losses on PWM harmonic losses through experimental observations and tests.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2022)
Article
Engineering, Multidisciplinary
Silvio Vaschetto, Zbigniew Gmyrek, Christoph Dobler, Gerd Bramerdorfer, Andrea Cavagnino
Summary: The use of interlocks in mass production of electrical machines can provide an affordable stacking solution for soft magnetic cores. However, this introduces additional conductive paths, leading to increased eddy current losses, as well as locally introduced mechanical stresses that cause degradation of the magnetic material properties. This article develops a reliable three-dimensional finite-element method model to analyze the impact of interlocks on flux and eddy current density distributions, as well as compute total iron losses. The numerical analyses are validated using interlaboratory measurements, revealing significant impact on core losses and material BH curve degradation.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2022)
Article
Engineering, Electrical & Electronic
Cesar Gallardo, Juan A. Tapia, Michele Degano, Hanafy Mahmoud
Summary: This article presents an accurate analytical model for synchronous reluctance machine, which considers the effect of stator slots on the performance of the machine. The model allows for fast prediction of torque waveforms and flux density, and provides correlations between electrical, magnetic, and geometrical characteristics of the machine, facilitating the design and optimization process.
IEEE TRANSACTIONS ON MAGNETICS
(2022)
Article
Energy & Fuels
Mostafa Ahmadi Darmani, Emir Poskovic, Fausto Franchini, Luca Ferraris, Andrea Cavagnino
Summary: This study explores the potential of various types of permanent magnet materials in building and characterizing multilayer magnets, as well as their prospective applications in electrical machines. Bonded magnet and hybrid magnet powders are used to construct double-layer samples, paving the way for building special electrical machines with complex magnetic structures.
IEEE TRANSACTIONS ON ENERGY CONVERSION
(2022)
Article
Energy & Fuels
Luis F. D. Bucho, Joao F. P. Fernandes, Marco Biasion, Silvio Vaschetto, Andrea Cavagnino
Summary: This work presents an experimental assessment of the influence of cryogenic cooling on a conventional induction motor. The study finds that when submerged in liquid nitrogen, the motor's performance parameters vary and it achieves higher efficiency and torque.
IEEE TRANSACTIONS ON ENERGY CONVERSION
(2022)
Article
Automation & Control Systems
Alvaro E. Hoffer, Roberto H. Moncada, Boris J. Pavez-Lazo, Juan A. Tapia, Lasse Laurila
Summary: This article proposes a simple method to calculate a current vector trajectory for the enhanced operation of a synchronous reluctance machine (SynRM) in an electric power generation system. It takes into account the magnetic saturation and cross magnetization effects, which affect the performance and torque capability of the machine. The proposed trajectory, based on the machine’s inductance characteristic, is shown to improve the torque capability by 5% compared to not considering saturation, as demonstrated through numerical simulations and experimental results.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2023)
Article
Chemistry, Multidisciplinary
Cesar Gallardo, Carlos Madariaga, Juan A. A. Tapia, Michele Degano
Summary: A discrete skew methodology is introduced to investigate the impact of skewing angle on electromagnetic torque in SynRM design. A novel approach is proposed to estimate the amplitude of each torque ripple component based on the skewing angle. The reduction factor for each harmonic component is derived in a general form, allowing for the determination of the overall torque ripple waveform. The proposed method is validated through the analysis of two SynRMs, achieving a torque ripple reduction of up to 70%.
APPLIED SCIENCES-BASEL
(2023)
Article
Chemistry, Multidisciplinary
Danilo Riquelme, Carlos Madariaga, Werner Jara, Gerd Bramerdorfer, Juan A. Tapia, Javier Riedemann
Summary: This paper addresses the importance of addressing stator-rotor misalignment, or eccentricity, in the design stage of modular permanent magnet synchronous machines. Finite element method is used to evaluate the static and dynamic eccentricity for different slot/pole combinations and compared with conventional PMSMs. The main findings suggest that eccentricity leads to severe radial forces and additional cogging torque harmonics, and the difference between modular PMSMs and conventional PMSMs lies in the value of slots per pole per phase.
APPLIED SCIENCES-BASEL
(2023)
Article
Engineering, Electrical & Electronic
Andres Escobar, Gonzalo Sanchez, Werner Jara, Carlos Madariaga, Juan A. Tapia, Javier Riedemann, Eduardo Reyes
Summary: Axial Flux Permanent Magnet (AFPM) machines with ironless rotors are attractive for low-speed applications due to their high power/weight ratio, high aspect ratio, and high efficiency. However, manufacturing tolerances can affect these machines, leading to freely moving magnets in the rotor structure. This study presents a statistical analysis of manufacturing tolerances in an AFPM machine with an ironless rotor, considering different magnet fault types. A computationally efficient superposition method is developed to analyze the cogging torque and rated torque under various tolerance combinations. The results reveal the significant impact of specific parameters on performance indicators and the considerable increase in cogging and ripple torque with inevitable combined tolerances.
Proceedings Paper
Computer Science, Interdisciplinary Applications
Cesar Gallardo, Carlos Madariaga, Vanesa Rodriguez-Merchan, Juan A. Tapia
Summary: This article presents a comparative study between a synchronous reluctance machine and a solid rotor induction machine, both operating at a mechanical speed of 20,000 r/min. The results show that the SynRM has lower core losses and higher overall efficiency, but higher torque ripple compared to the SR-IM. The SR-IM, on the other hand, has significantly lower mechanical stress and can withstand higher operation speeds without compromising rotor structure integrity.
2022 IEEE ANDESCON
(2022)
Proceedings Paper
Engineering, Electrical & Electronic
Zbigniew Gmyrek, Silvio Vaschetto, Andrea Cavagnino
Summary: The paper presents a new method to estimate surface losses caused by non-sinusoidal spatial distribution of the magnetic field on laminated rotor cores. The rotating magnetic field in the air gap of AC electrical machines causes additional losses, especially in high-speed applications. A simplified finite element method model was developed to simulate the mechanism of surface loss generation, with results critically discussed and compared with other research works.
2021 IEEE INTERNATIONAL ELECTRIC MACHINES & DRIVES CONFERENCE (IEMDC)
(2021)
Article
Energy & Fuels
Federica Graffeo, Silvio Vaschetto, Alessio Miotto, Fabio Carbone, Alberto Tenconi, Andrea Cavagnino
Summary: Thermal analysis is crucial in electrical machine design, particularly for harsh working conditions. Brake-by-wire systems are challenging applications for electric machines, requiring repeated operation in high overload conditions while maintaining temperature balance. This study aims to find a thermal model suitable for electric machines used in brake-by-wire systems to accurately predict temperature variations.