Article
Computer Science, Artificial Intelligence
Caitong Yue, P. N. Suganthan, Jing Liang, Boyang Qu, Kunjie Yu, Yongsheng Zhu, Li Yan
Summary: This paper proposes a multimodal multiobjective differential evolution algorithm to solve the problem of many-to-one mappings in multiobjective optimization. The proposed method takes into account the diversity in both decision and objective space, and changes the way of calculating crowding distance to improve solution diversity.
SWARM AND EVOLUTIONARY COMPUTATION
(2021)
Article
Computer Science, Artificial Intelligence
Ting Zhou, Zhongbo Hu, Qinghua Su, Wentao Xiong
Summary: This paper proposes a novel multimodal multiobjective differential evolution algorithm (MMOcDE) that can locate multiple high quality equivalent Pareto optimal sets and obtain a uniformly distributed Pareto front simultaneously.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Bilal, Millie Pant, Vaclav Snasel
Summary: This study proposes a fuzzy C-means adaptive differential evolution (FCADE) algorithm to solve complex problems in water distribution networks (WDN). The algorithm, integrated with the simulation software EPANET, demonstrates strong solving capabilities.
Article
Automation & Control Systems
Xiao-Fang Liu, Jun Zhang, Jun Wang
Summary: This article presents a cooperative differential evolution algorithm with an attention-based prediction strategy for dynamic multiobjective optimization. Multiple populations are used to optimize multiple objectives and find subparts of the Pareto front. The algorithm achieves a balanced approximation of the Pareto front and adapts to changes in the environment by using a new attention-based prediction strategy. Experimental results demonstrate the superiority of the proposed method to state-of-the-art algorithms.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Computer Science, Information Systems
Zhiqiang Zeng, Xiangyu Zhang, Zhiyong Hong
Summary: A novel constraint handling technique (CHT) that fuses two rankings is proposed in this paper, addressing the tradeoff between objective functions and constraints in constrained multiobjective optimization algorithms. Based on this CHT, a constrained multiobject differential evolution algorithm is proposed which combines four mutation operations to generate high-quality offspring. Experimental results demonstrate that the proposed algorithm outperforms eight state-of-the-art algorithms in five test suites.
INFORMATION SCIENCES
(2023)
Article
Chemistry, Multidisciplinary
Van-Tinh Nguyen, Vu-Minh Tran, Ngoc-Tam Bui
Summary: In this study, a modified version of the ISADE algorithm is proposed, which applies the Gauss distribution for the mutation procedure to enhance the diversity of the population. Simulation results show that the suggested algorithm performs exceptionally well compared to other reference algorithms in terms of converging speed and consistency of optimal solutions.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Naili Luo, Wu Lin, Genmiao Jin, Changkun Jiang, Jianyong Chen
Summary: In this paper, a novel genetically hybrid differential evolution strategy (GHDE) for recombination in MOEA/Ds is proposed to enhance search capability by introducing two composite operator pools. Through adaptive parameter tuning and fitness-rate-rank-based multiarmed bandit (FRRMAB), the best operator pool is selected, demonstrating the superiority of MOEA/D-GHDE in multiobjective optimization problems.
Article
Computer Science, Information Systems
Shuai Wang, Aimin Zhou, Bingdong Li, Peng Yang
Summary: The article proposes a new differential evolution algorithm (RMDE) that improves the search performance for multiobjective optimization problems by sampling guiding solutions from regularity models.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Yupeng Han, Hu Peng, Changrong Mei, Lianglin Cao, Changshou Deng, Hui Wang, Zhijian Wu
Summary: This paper proposes a new multistrategy multiobjective differential evolutionary algorithm, RLMMDE, to solve the exploration and exploitation dilemma in multiobjective optimization problems (MOPs). The algorithm utilizes a multistrategy and multicrossover DE optimizer, an adaptive reference point activation mechanism based on RL, and a reference point adaptation method. Experimental results show that RLMMDE outperforms some advanced MOEAs on benchmark test suites and practical mixed-variable optimization problems.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Jing Liang, Xuanxuan Ban, Kunjie Yu, Kangjia Qiao, Boyang Qu
Summary: This paper presents a constrained multiobjective differential evolution algorithm with an infeasible-proportion control mechanism, which addresses the handling of conflicting objectives and constraints through cooperative strategies and infeasible-proportion control. Experimental results demonstrate that the proposed algorithm outperforms or is at least comparable to existing constrained multiobjective optimization methods on various benchmark test functions.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Xiaobing Yu, Chenliang Li, Gary G. Yen
Summary: This paper presents a multiobjective optimization problem for path planning in a three-dimensional terrain disaster scenario. A differential evolution algorithm based on knee point is proposed to efficiently generate smooth paths and identify optimal solutions. Experimental results confirm the superiority of the algorithm.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Information Systems
Qinghua Gu, Siping Huang, Qian Wang, Xuexian Li, Di Liu
Summary: This article proposes a Chaotic Differential Evolution and Symmetric Direction Sampling (CDE-SDS) method for large-scale multiobjective optimization. The CDE-SDS method utilizes chaotic differential evolution strategy to accelerate convergence and employs symmetric direction sampling strategy to explore the high-dimensional decision space. Experimental results show that CDE-SDS outperforms seven compared algorithms in terms of diversity and convergence under limited function evaluations.
INFORMATION SCIENCES
(2023)
Article
Operations Research & Management Science
Luciano Ferreira Cruz, Flavia Bernardo Pinto, Lucas Camilotti, Angelo Marcio Oliveira Santanna, Roberto Zanetti Freire, Leandro dos Santos Coelho
Summary: The study proposes a hybrid approach to optimize 3D printing technology parameters for solving engineering problems and decision-making support, with the multiobjective differential evolution algorithm showing superior performance.
ANNALS OF OPERATIONS RESEARCH
(2022)
Article
Automation & Control Systems
Vikas Palakonda, Jae-Mo Kang
Summary: This article proposes a preference-inspired differential evolution algorithm for multi and many-objective optimization, which effectively deals with a wide range of problems. The algorithm generates individuals with good convergence and distribution properties by utilizing a preference-inspired mutation operator and determining local knee points based on a clustering method. Experimental results demonstrate its superior performance compared to eight state-of-the-art algorithms on 35 benchmark problems.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Mathematics, Interdisciplinary Applications
Jing Liu, Qixing Chen, Xiaoying Tian
Summary: This paper proposes an illustration art design model based on operator and clustering optimization genetic algorithm, as well as a multiobjective optimization genetic algorithm with complex constraints based on group classification. By improving the genetic algorithm, the search efficiency is increased and the complexity of the algorithm is optimized.
Article
Mathematics, Interdisciplinary Applications
Matheus Henrique Dal Molin Ribeiro, Ramon Gomes da Silva, Gabriel Trierweiler Ribeiro, Viviana Cocco Mariani, Leandro dos Santos Coelho
Summary: Efficient models for short-term load forecasting in electricity distribution and generation systems are crucial for companies' energetic planning. In this study, an ensemble learning model based on dual decomposition approach, machine learning models and hyperparameters optimization is proposed. The model successfully decomposes the time series and handles the non-linearities, and achieves accurate load forecasting results with reduced errors.
CHAOS SOLITONS & FRACTALS
(2023)
Article
Energy & Fuels
Stefano Frizzo Stefenon, Laio Oriel Seman, Viviana Cocco Mariani, Leandro dos Santos Coelho
Summary: The cost of electricity and gas has a direct impact on people's everyday routines, but the value of electricity is closely related to spot market prices, which can increase in winter due to higher energy demand. Existing models for forecasting energy costs are not robust enough due to competition, seasonal changes, and other variables. This study proposes combining seasonal and trend decomposition using LOESS and Facebook Prophet methodologies to improve the accuracy of analyzing time series data on Italian electricity spot prices.
Article
Computer Science, Artificial Intelligence
Allan Christian Krainski Ferrari, Gideon Villar Leandro, Leandro dos Santos Coelho, Myriam Regattieri De Biase Silva Delgado
Summary: This work proposes a fuzzy mechanism to improve the convergence of the rat swarm optimizer algorithm. The proposed fuzzy model uses the normalized fitness and population diversity as input. The results show that the fuzzy mechanism improves convergence and is competitive with other metaheuristics.
JOURNAL OF INTELLIGENT & FUZZY SYSTEMS
(2023)
Article
Computer Science, Information Systems
Andre Armstrong Janino Cizotto, Rodrigo Clemente Thom de Souza, Viviana Cocco Mariani, Leandro dos Santos Coelho
Summary: The objective of this study is to validate the use of deep neural networks (DNNs) for segmenting and classifying web elements. A dataset of 2200 images representing 10 distinct classes was created using screenshots of real web pages. The study contributes by validating classification-only convolutional neural networks (CNNs) with the support of Class Activation Mapping (CAM), a weakly-supervised semantic segmentation technique. The best-performing model achieved a final accuracy rating of 95.71%, but improvements are still needed on the dataset and architecture for real-time dynamic web page building.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Chemistry, Analytical
Anne Carolina Rodrigues Klaar, Stefano Frizzo Stefenon, Laio Oriel Seman, Viviana Cocco Mariani, Leandro dos Santos Coelho
Summary: Insulators installed outdoors are prone to accumulation of contaminants, causing increased conductivity and leakage current, eventually leading to flashover. To enhance power system reliability, it is possible to predict fault development and potential shutdown by evaluating the increase in leakage current. This paper proposes a method, optimized EWT-Seq2Seq-LSTM with attention, which combines empirical wavelet transform (EWT) to reduce non-representative variations and the attention mechanism with LSTM recurrent network for prediction. The model achieved a 10.17% lower mean square error (MSE) compared to standard LSTM and a 5.36% lower MSE compared to the model without optimization, demonstrating the effectiveness of the attention mechanism and hyperparameter optimization.
Article
Chemistry, Analytical
Andressa Borre, Laio Oriel Seman, Eduardo Camponogara, Stefano Frizzo Stefenon, Viviana Cocco Mariani, Leandro dos Santos Coelho
Summary: In this paper, the issue of predicting electrical machine failures by predicting possible anomalies in the data is addressed through time series analysis. The dataset is used to train a hybrid CNN-LSTM architecture, which employs quantile regression to manage uncertainties in the data. The results show that this approach outperforms traditional reference models, making it beneficial for companies to optimize maintenance schedules and improve the performance of their electric machines.
Article
Chemistry, Analytical
Guilherme Augusto Silva Surek, Laio Oriel Seman, Stefano Frizzo Stefenon, Viviana Cocco Mariani, Leandro dos Santos Coelho
Summary: This paper aims to evaluate and map the current scenario of human actions in red, green, and blue videos using deep learning models. A semi-supervised learning approach is employed to evaluate a residual network (ResNet) and a vision transformer architecture (ViT). The results obtained using a bi-dimensional ViT structure demonstrated great performance in human action recognition, achieving an accuracy of 96.7% on the HMDB51 dataset.
Article
Chemistry, Analytical
Matheus Henrique Dal Molin Ribeiro, Ramon Gomes da Silva, Jose Henrique Kleinubing Larcher, Andre Mendes, Viviana Cocco Mariani, Leandro dos Santos Coelho
Summary: This paper proposes a new hybrid framework combining STACK ensemble learning and a JADE algorithm for nonlinear system identification. The model performs well in decoding EEG signals, achieving an average explanation of 94.50% and 67.50% of data variability, and outperforms other methods in terms of accuracy.
Article
Chemistry, Analytical
Stefano Frizzo Stefenon, Laio Oriel Seman, Nemesio Fava Sopelsa Neto, Luiz Henrique Meyer, Viviana Cocco Mariani, Leandro dos Santos Coelho
Summary: This paper presents a novel hybrid method for fault prediction based on the time series of leakage current of contaminated insulators. The proposed CFRW-GMDH method, with a root-mean-squared error of 3.44x10(-12), outperformed other models in fault prediction. This approach can provide power utilities with a reliable tool for monitoring insulator health and predicting failures, thereby enhancing the reliability of the power supply.
