Article
Computer Science, Information Systems
Kareem Kamal A. Ghany, Amr Mohamed AbdelAziz, Taysir Hassan A. Soliman, Adel Abu El-Magd Sewisy
Summary: This paper proposes a data clustering method called WOATS, which combines Whale Optimization Algorithm (WOA) with Tabu Search (TS). WOATS uses an objective function inspired by partitional clustering to maintain the quality of clustering solutions. Experimental results show that WOATS outperforms other recent SI methods on different datasets, demonstrating its efficiency.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Bin Li, Ziping Wei, Jingjing Wu, Shuai Yu, Tian Zhang, Chunli Zhu, Dezhi Zheng, Weisi Guo, Chenglin Zhao, Jun Zhang
Summary: Evolutionary computation has achieved impressive results in solving complex problems, but there is no theoretical guarantee for reaching the global optimum. To address this challenge, researchers have proposed an evolutionary computation framework called EVOLER, aided by machine learning, which enables theoretically guaranteed global optimization of complex non-convex problems. This is achieved by learning a low-rank representation of the problem and exploring a small attention subspace using evolutionary computation methods to reliably avoid local optima.
NATURE MACHINE INTELLIGENCE
(2023)
Review
Computer Science, Artificial Intelligence
Feng Qin, Azlan Mohd Zain, Kai-Qing Zhou
Summary: This article systematically reviews the harmony search (HS) algorithm and its variants from three aspects: describing the basic HS principle, discussing the impact of HS improvement on algorithm performance, and analyzing the characteristics and applications of HS variants. It is found that the improvement of HS mainly focuses on parameter enhancement and the integration with other metaheuristic algorithms, providing future directions for enhancing HS.
SWARM AND EVOLUTIONARY COMPUTATION
(2022)
Article
Multidisciplinary Sciences
Abiodun M. Ikotun, Absalom E. Ezugwu
Summary: This paper proposes a hybrid clustering method that combines the symbiotic organisms search algorithm with K-Means for generating optimum initial cluster centroids. The results from extensive experimentation on multiple datasets demonstrate improved performance of the hybrid algorithm for automatic clustering.
Article
Multidisciplinary Sciences
Mohammad Reza Sharifi, Saeid Akbarifard, Kourosh Qaderi, Mohamad Reza Madadi
Summary: The study employed five robust evolutionary algorithms for optimal operation of a multi-reservoir system and found that the MSA algorithm performed the best in terms of objective function value, CPU run-time, and convergence rate. This suggests that the application of robust EAs, particularly the MSA algorithm, is recommended to improve the operation policies of multi-reservoir systems.
SCIENTIFIC REPORTS
(2021)
Article
Computer Science, Artificial Intelligence
Junhao Huang, Bing Xue, Yanan Sun, Mengjie Zhang, Gary G. Yen
Summary: Neural architecture search (NAS) is a popular research topic in deep learning community due to its potential in automating the construction of deep models. Among various NAS approaches, evolutionary computation (EC) stands out for its capability of gradient-free search. However, most current EC-based NAS approaches have the limitation of discrete evolution, making it difficult to handle the number of filters for each layer flexibly. Additionally, EC-based NAS methods are criticized for their inefficiency in performance evaluation, often requiring full training of hundreds of candidate architectures. This work proposes a split-level particle swarm optimization (PSO) approach to address these issues and achieves superior performance on image classification benchmarks.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Automation & Control Systems
Xiao Feng, Hisayoshi Muramatsu, Seiichiro Katsura
Summary: Periodic disturbances during repetitive operations affect machine works precision; it becomes challenging to compensate for frequency-varying disturbances; a differential evolutionary algorithm is proposed to optimize design parameters for better compensation.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2021)
Review
Computer Science, Information Systems
Guangquan Li, Ting Wang, Qi Chen, Peng Shao, Naixue Xiong, Athanasios Vasilakos
Summary: This paper reviews the existing research on heuristic algorithms based on Particle Swarm Optimization (PSO) in association rule mining (ARM), and describes different types of ARM and their evaluation metrics. Furthermore, the current status of improvement in PSO algorithms and the applications of PSO-based ARM algorithms are discussed, along with proposed further research directions.
Article
Computer Science, Information Systems
Bingwei Gao, Wei Shen, Hao Guan, Lintao Zheng, Wei Zhang
Summary: This paper proposes an improved evolutionary sparrow search algorithm to address the issues of poor global search ability, weak local development ability, and easy fall into the local optimal solution. By introducing the chaotic map, random search ability, and mutation evolution operation, the algorithm shows superior performance in optimization ability, robustness, convergence ability, and optimization trajectory. The algorithm is also applied to parameter identification and control strategies, demonstrating its advantages in practical engineering applications.
Article
Computer Science, Artificial Intelligence
Junbo Lian, Guohua Hui
Summary: This paper introduces the Human Evolutionary Optimization Algorithm (HEOA), which is a metaheuristic algorithm inspired by human evolution. The algorithm divides the global search process into two distinct phases and uses unique search strategies. Comparative analysis with other algorithms demonstrates the effectiveness of HEOA in approximating optimal solutions for complex global optimization problems.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Mathematical & Computational Biology
Xuepeng Zheng, Bin Nie, Jiandong Chen, Yuwen Du, Yuchao Zhang, Haike Jin
Summary: The paper proposes an improved particle swarm optimization algorithm combined with double-chaos search (DCS-PSO). DCS-PSO narrows the search space and enhances population diversity by incorporating a double-chaos search mechanism and a logistic map. Experimental results demonstrate that DCS-PSO achieves better convergence accuracy and speed in most cases.
MATHEMATICAL BIOSCIENCES AND ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Malik Braik, Hussein Al-Zoubi, Mohammad Ryalat, Alaa Sheta, Omar Alzubi
Summary: Crow Search Algorithm (CSA) is a promising meta-heuristic method that mimics the intelligent behavior of crows in nature. By combining it with Particle Swarm Optimization (PSO), the Memory based Hybrid CSA (MHCSA) achieves a stronger diversity ability and a better balance between exploration and exploitation, effectively overcoming the early convergence and imbalance issues. Test results have demonstrated the superiority of MHCSA over other methods in terms of accuracy and stability.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Computer Science, Artificial Intelligence
Mohammed Qaraad, Souad Amjad, Nazar K. Hussein, M. A. Farag, Seyedali Mirjalili, Mostafa A. Elhosseini
Summary: The Particle Swarm Optimization technique is widely used but has limitations. To address these issues, a novel local search technique called QPSOL is proposed, which aims to increase diversity and achieve a closer balance between exploration and exploitation. QPSOL incorporates dynamic optimization strategy and quadratic interpolation to enhance the performance of PSO.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Computer Science, Interdisciplinary Applications
Maliheh Abbaszadeh, Saeed Soltani-Mohammadi, Ali Najah Ahmed
Summary: This article introduces the application of the support vector classifier in geological modeling and proposes an improved method based on particle swarm optimization to select the best model parameters. Through the application in the modeling process of the Iju porphyry copper deposit, the effectiveness and superiority of this method are demonstrated.
COMPUTERS & GEOSCIENCES
(2022)
Article
Multidisciplinary Sciences
Pham Vu Hong Son, Nghiep Trinh Nguyen Dang
Summary: The study introduces a hybrid multi-verse optimizer model (hDMVO) that combines the multi-verse optimizer (MVO) and the sine cosine algorithm (SCA) to solve the discrete time-cost trade-off problem (DTCTP). The optimality of the algorithm is evaluated using 23 benchmark test functions, demonstrating its competitiveness with other algorithms. The performance of hDMVO is further evaluated using four benchmark test problems, showing its superiority in time-cost optimization for large-scale and complex projects compared to previous algorithms.
SCIENTIFIC REPORTS
(2023)
Article
Automation & Control Systems
Tiange Wang, Zijun Zhang, Kwok-Leung Tsui
Summary: This article proposes a deep generative approach for detecting foreign objects on railway tracks. The approach involves training a model using an autoencoder and discriminator, detecting abnormal images based on anomaly scores obtained from the trained autoencoder, and filtering normal areas to highlight abnormal areas for foreign object detection.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Green & Sustainable Science & Technology
Hong Liu, Zijun Zhang
Summary: This paper proposes a novel data-driven modeling framework for short-term wind power prediction (WPP) that incorporates SCADA data of both high and low sampling resolutions. Extensive computational experiments based on real wind farm data demonstrate the effectiveness of the proposed framework and identify suitable machine learning techniques for feature selection and prediction. The framework achieves state-of-the-art performance by surpassing classical data-driven and recent deep learning based WPP methods considered in this study.
JOURNAL OF CLEANER PRODUCTION
(2022)
Article
Computer Science, Information Systems
Li Zhuang, Haoyang Qi, Tiange Wang, Zijun Zhang
Summary: A two-stage method using deep learning is developed for automating the inspection of major components in railway tracks. The method achieves accurate detection through two stages and utilizes a hybrid model based on domain knowledge to enhance performance. Field collected railway images are used to validate the effectiveness of the proposed method, which is shown to outperform other methods in the task of railway track inspection.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Computer Science, Interdisciplinary Applications
Haike Qiao, Zijun Zhang, Qin Su
Summary: This study investigates the adoption of blockchain technology (BCT) for verifying product recycling information (VPRI) in the presence of original and green consumers. Three modes of the manufacturer are analyzed: no adoption of BCT, adopting own BCT, and adopting a third-party BCT platform for VPRI. The results show that the manufacturer's adoption mode depends on the scaling cost and the proportion of BCT verified to all recycling information. Furthermore, the study finds that BCT adoption can have negative environmental impacts, especially when the carbon emission rate of recycled products is higher.
COMPUTERS & INDUSTRIAL ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Zhong Zheng, Zijun Zhang
Summary: In this article, a stochastic recurrent encoder-decoder neural network (SREDNN) is developed for generative multistep probabilistic wind power predictions. The SREDNN considers latent random variables in its recurrent structures and provides two critical advantages compared to conventional RNN-based methods: it models wind power distribution using an infinite Gaussian mixture model (IGMM) and describes complex patterns across wind speed and power sequences by updating hidden states in a stochastic way. Computational experiments demonstrate the advantages and effectiveness of the SREDNN for wind power prediction.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Luoxiao Yang, Long Wang, Zijun Zhang
Summary: The study presents a novel method, called DITU-net, for automating wind power curve (WPC) modeling without data pre-processing. The proposed approach incorporates a data-synthesis-informed-training (DIT) process to generate diverse training samples, which are then used to train a U-net model for the generation of neat WPC. Additionally, a pixel mapping and correction process is developed to derive a mathematical form depicting the neat WPC. The method eliminates the need for data preprocessing and achieves superior performance compared to classical WPC modeling methods.
IEEE TRANSACTIONS ON POWER SYSTEMS
(2023)
Article
Automation & Control Systems
Zhenling Mo, Zijun Zhang, Qiang Miao, Kwok-Leung Tsui
Summary: This article introduces a new dynamic bandit tree (DBT) algorithm to help achieve more adaptive filters and reduce the burden of parameter tuning in frequency band searching. By optimizing the boundaries of Meyer wavelet filters, this method can better identify demodulated fault frequencies and outperform other optimization algorithms and fault diagnosis methods in tests.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2023)
Article
Computer Science, Artificial Intelligence
Lanlan Zheng, Xin Liu, Feng Wu, Zijun Zhang
Summary: This paper addresses the two-dimensional shelf space allocation problem (2DSSAP) in the retail field by proposing a data-driven model assisted hybrid genetic algorithm (DMA-HGA). The proposed DMA-HGA applies an improved genetic algorithm (GA) to optimize solutions and a two-stage search assistance module to enhance the search process. Experimental results demonstrate that the DMA-HGA outperforms benchmarking methods in terms of solution quality and accuracy. The extended discussion of parameters also provides valuable management insights for the 2DSSAP.
SWARM AND EVOLUTIONARY COMPUTATION
(2023)
Article
Energy & Fuels
Zicheng Fei, Zijun Zhang, Fangfang Yang, Kwok-Leung Tsui
Summary: This paper proposes a new approach to predict the remaining useful life (RUL) of lithium-ion batteries using a limited number of incomplete cycles. The attention-assisted temporal convolutional memory-augmented network (ATCMN) is developed to achieve accurate and rapid RUL prediction under this challenging scenario. Experimental results demonstrate the effectiveness and generalizability of the proposed ATCMN compared to state-of-the-art methods.
JOURNAL OF ENERGY STORAGE
(2023)
Article
Computer Science, Artificial Intelligence
Zhong Zheng, Zijun Zhang
Summary: This paper proposes a temporal convolutional recurrent autoencoder framework for more effective time series compression. Experimental results show that the proposed method outperforms benchmarking models in terms of lower reconstruction errors with the same compression ratio, indicating its promising potential for various applications involving long time series data.
APPLIED SOFT COMPUTING
(2023)
Article
Thermodynamics
Hong Liu, Luoxiao Yang, Bingying Zhang, Zijun Zhang
Summary: This paper presents a pioneering attempt of studying a two-channel deep network modeling method for wind power predictions that leverages both wind farm data and farm geoinformation. Through comprehensive computational experiments and comparison with benchmarking models, the value of this modeling approach is confirmed, achieving a new state-of-the-art prediction performance.
Article
Engineering, Civil
Tiange Wang, Zijun Zhang, Kwok-Leung Tsui
Summary: Predicting vehicle crashes accurately is challenging due to the dominance of non-crash data. Existing studies have limitations in predicting crashes in advance. This paper proposes a crash alarm model for vehicles (CAMV) that uses vehicle operational data. The proposed model utilizes a non-crash learning block (NCLB) and a coarse-to-fine strategy to generate and calibrate alarms. Experimental results show the effectiveness of CAMV and its superiority over benchmarking methods.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Hong Liu, Zijun Zhang
Summary: This paper presents a pioneering study on using numerical weather predictions to predict future renewable power output while preserving data privacy. It introduces a novel bi-party engaged data-driven modeling framework (BEDMF) that learns local and global latent features and captures spatial-temporal patterns among multiple sites. Experimental results show that the BEDMF achieves at least 3% improvement on average compared to famous baselines.
IEEE TRANSACTIONS ON POWER SYSTEMS
(2023)
Article
Green & Sustainable Science & Technology
Zhong Zheng, Luoxiao Yang, Zijun Zhang
Summary: In this article, a conditional variational autoencoder based method is proposed for the probabilistic wind power curve modeling task. Experimental results show that the proposed method has high performance and reliability in simulating wind power curves.
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
(2023)
Article
Energy & Fuels
Zicheng Fei, Zijun Zhang, Fangfang Yang, Kwok-Leung Tsui
Summary: This study develops a deep learning powered method for rapid lifetime classification of lithium-ion batteries using limited early-cycle data. The method integrates spatial, temporal, and physical battery information, extracts high-level latent features, and classifies batteries accurately.
Article
Construction & Building Technology
Samiran Khorat, Debashish Das, Rupali Khatun, Sk Mohammad Aziz, Prashant Anand, Ansar Khan, Mattheos Santamouris, Dev Niyogi
Summary: Cool roofs can effectively mitigate heatwave-induced excess heat and enhance thermal comfort in urban areas. Implementing cool roofs can significantly improve urban meteorology and thermal comfort, reducing energy flux and heat stress.
ENERGY AND BUILDINGS
(2024)
Article
Construction & Building Technology
Qi Li, Jiayu Chen, Xiaowei Luo
Summary: This study focuses on the vertical wind conditions as a main external factor that limits the energy assessment of high-rise buildings in urban areas. Traditional tools for energy assessment of buildings use a universal vertical wind profile estimation, without taking into account the unique wind speed in each direction induced by the various shapes and configurations of buildings in cities. To address this limitation, the study developed an omnidirectional urban vertical wind speed estimation method using direction-dependent building morphologies and machine learning algorithms.
ENERGY AND BUILDINGS
(2024)
Article
Construction & Building Technology
Xiaojun Luo, Lamine Mahdjoubi
Summary: This paper presents an integrated blockchain and machine learning-based energy management framework for multiple forms of energy allocation and transmission among multiple domestic buildings. Machine learning is used to predict energy generation and consumption patterns, and the proposed framework establishes optimal and automated energy allocation through peer-to-peer energy transactions. The approach contributes to the reduction of greenhouse gas emissions and enhances environmental sustainability.
ENERGY AND BUILDINGS
(2024)
Article
Construction & Building Technology
Ying Yu, Yuanwei Xiao, Jinshuai Chou, Xingyu Wang, Liu Yang
Summary: This study proposes a dual-layer optimization design method to maximize the energy sharing potential, enhance collaborative benefits, and reduce the storage capacity of building clusters. Case studies show that the proposed design significantly improves the performance of building clusters, reduces energy storage capacity, and shortens the payback period.
ENERGY AND BUILDINGS
(2024)
Article
Construction & Building Technology
Felix Langner, Weimin Wang, Moritz Frahm, Veit Hagenmeyer
Summary: This paper compares two main approaches to consider uncertainties in model predictive control (MPC) for buildings: robust and stochastic MPC. The results show that compared to a deterministic MPC, the robust MPC increases the electricity cost while providing complete temperature constraint satisfaction, while the stochastic MPC slightly increases the electricity cost but fulfills the thermal comfort requirements.
ENERGY AND BUILDINGS
(2024)
Article
Construction & Building Technology
Somil Yadav, Caroline Hachem-Vermette
Summary: This study proposes a mathematical model to evaluate the performance of a Double Skin Facade (DSF) system and its impact on indoor conditions. The model considers various design parameters and analyzes their effects on the system's electrical output and room temperature.
ENERGY AND BUILDINGS
(2024)
Article
Construction & Building Technology
Ruijun Chen, Holly Samuelson, Yukai Zou, Xianghan Zheng, Yifan Cao
Summary: This research introduces an innovative resilient design framework that optimizes building performance by considering a holistic life cycle perspective and accounting for climate projection uncertainties. The study finds that future climate scenarios significantly impact building life cycle performance, with wall U-value, windows U-value, and wall density being major factors. By using ensemble learning and optimization algorithms, predictions for carbon emissions, cost, and indoor discomfort hours can be made, and the best resilient design scheme can be selected. Applying this framework leads to significant improvements in building life cycle performance.
ENERGY AND BUILDINGS
(2024)