Article
Mathematics, Interdisciplinary Applications
Xinyao Wang, Huanwen Jiang, Guosheng Han
Summary: This article introduces a method for multifractal analysis of nonlinear time series and applies it to the multifractal analysis of urban and suburban areas. The study finds that both urban and suburban systems exhibit multifractality, with the urban system showing stronger multifractality, particularly in spring and winter.
CHAOS SOLITONS & FRACTALS
(2023)
Article
Chemistry, Physical
Xiaoshuang Shi, Cong Zhang, Yongchen Liang, Jinqian Luo, Xiaoqi Wang, Ying Feng, Yanlin Li, Qingyuan Wang, Abd El-Fatah Abomohra
Summary: The study demonstrates that the use of geopolymer concrete can significantly reduce CO2 emissions and has minimal impact on compressive strength. Among the key parameters, sodium hydroxide concentration has the greatest impact on compressive strength, while the alkali activator solution to fly ash ratio has the greatest impact on CO2 emissions.
Article
Green & Sustainable Science & Technology
Lei Wu, Jiao Liu, Jun Zhou, Qiuli Zhang, Yonghui Song, Shuai Du, Wei Tian
Summary: This study investigates the use of corncob as a hydrogen donor to improve the yield and quality of tar produced through microwave pyrolysis of low-rank coal. The results show that the optimal conditions yield a tar with a maximum yield of 8.85% and a lighter and more valuable structure.
JOURNAL OF CLEANER PRODUCTION
(2022)
Article
Computer Science, Information Systems
Ruijin Wang, Xikai Pei, Juyi Zhu, Zhiyang Zhang, Xin Huang, Jiayi Zhai, Fengli Zhang
Summary: This paper proposes a model fusion-based time series forecasting method to improve the accuracy and efficiency of predictions using multivariate grey model and artificial fish swarm algorithm. Two fusion models based on data decomposition and weighted summation achieve good prediction results in different scenarios.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Hailin Li, Tian Du
Summary: This study proposes an MTS clustering method based on a component relationship network (CRN), which consists of a multi-relationship network (MRN) mapping an MTS dataset and utilizing non-negative matrix factorization. By incorporating an improved penalty-coefficient dynamic time-warping algorithm, the method effectively measures the similarity between asynchronous MTS data and improves the accuracy and quality of clustering. Through experiments, it is demonstrated that the proposed method outperforms other clustering methods by considering component correlations and parameter optimization.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Social Sciences, Interdisciplinary
Qun Wang, Kai Huang, Mark Goh, Zeyu Jiao, Guozhu Jia
Summary: Smart data selection improves decision-making efficiency by quickly identifying valuable information from initial data. This study introduces a modified Decision-Making Trial and Evaluation Laboratory (DEMATEL) method based on objective data grey relational analysis (GRA) to enhance the analysis of time-series data. The results of applying this method to predict the remaining useful life (RUL) of aircraft engines indicate its accuracy and potential applications.
Article
Computer Science, Information Systems
Nimisha Ghosh, Indrajit Banerjee
Summary: This study focuses on real-time monitoring of Freezing of Gait in Parkinson's disease patients using wearable acceleration sensors, and proposes a FoG detection method based on Grey Relational Analysis. An ensemble learning approach is also illustrated. The simulation results demonstrate that the proposed methods have superior accuracy compared to other machine learning techniques.
INTERNET OF THINGS
(2021)
Article
Thermodynamics
Kuwar Mausam, Ashutosh Pare, Subrata Kumar Ghosh, A. K. Tiwari
Summary: The thermal performance and greenhouse gas emission of a flat plate collector-based solar energy harvesting system (SEHs) were experimentally investigated using Cu-MWCNTs/water hybrid nanofluid. Grey relational analysis (GRA) was used to verify and optimize the experiment's output. An L27 orthogonal array was employed to design the experimental data set with three different levels of flow rate, intensity, and inclination angle. GRA ranked flow rate as the most significant factor, followed by intensity and inclination angle. The use of hybrid nanofluid yielded a maximum instantaneous efficiency of 68.7% at 1.5 lpm flow rate, 400 W/m2 intensity, and 25 degrees inclination angle, while reducing greenhouse gas emissions by 21.42 kg per year.
THERMAL SCIENCE AND ENGINEERING PROGRESS
(2023)
Article
Computer Science, Information Systems
Qinpei Zhao, Guangda Yang, Kai Zhao, Jiaming Yin, Weixiong Rao, Lei Chen
Summary: This article proposes a methodology to measure the upper limit of predictability for multivariate time series with multivariate constraint relations. The key of the methodology is a novel entropy named Multivariate Constraint Sample Entropy (McSE) that incorporates the multivariate constraint relations for better predictability. The authors conducted a systematic evaluation over eight datasets and compared existing methods with their proposed predictability, finding that their method achieved higher predictability. They also discovered that forecasting algorithms that capture the multivariate constraint relation information can achieve higher accuracy, confirming the importance of multivariate constraint relations for predictability.
IEEE TRANSACTIONS ON BIG DATA
(2023)
Article
Thermodynamics
Honglin Lv, Xueye Chen, Xiangyang Wang, Xiangwei Zeng, Yongbiao Ma
Summary: This paper presents a multi-objective optimization method for a micromixer with Cantor fractal baffles using grey relational analysis. The comprehensive performance of the micromixer is evaluated through two aspects. Design variables are selected, and an orthogonal experiment table is obtained. The optimized micromixer shows significant improvements in both mixing index and Poiseuille number.
INTERNATIONAL JOURNAL OF HEAT AND MASS TRANSFER
(2022)
Article
Forestry
Ravita, Sunita Rawat, Harish Singh Ginwal, Santan Barthwal
Summary: This study aims to select salt-tolerant Eucalyptus clones for cultivation in high saline conditions. Through Grey Relational Analysis and Multiple Attribute Decision-Making model, the best performing clones under salt stress were chosen, providing a screening basis for growing Eucalyptus plants in salt-affected locations and promoting forest sustainability.
JOURNAL OF SUSTAINABLE FORESTRY
(2023)
Article
Green & Sustainable Science & Technology
Zhaojun Yang, Xiaoting Guo, Jun Sun, Yali Zhang
Summary: The study uses mathematical analyses to identify critical factors that enterprises need to focus on for sustainable supply chains, guiding them to make more sustainable decisions with limited resources. Future research may enhance the generalizability of findings by validating results across multiple countries.
ENVIRONMENT DEVELOPMENT AND SUSTAINABILITY
(2021)
Article
Engineering, Mechanical
Lan Wang, Nan Li, Ming Xie, Lifeng Wu
Summary: To develop a generalized nonlinear multivariable grey model, we nonlinearize the traditional GM(1,N) model and call it NGM(1,N). NGM(1,N) and its response function include unidentified nonlinear functions that map the data into a better representational space. We establish the original optimization problem with linear equality constraints for parameter estimation and solve it using the Lagrange multiplier method and the standard dualization method. The results show that LDNGM(1,N) outperforms other multivariate grey models in terms of generalization performance, and the duality theory and framework with kernel learning provide guidance for future research on multivariate grey models.
NONLINEAR DYNAMICS
(2023)
Article
Computer Science, Artificial Intelligence
Yaoguo Dang, Yifan Zhang, Junjie Wang
Summary: To address the problem of the grey multivariate prediction model's inability to accurately simulate systems with periodic oscillations, a novel multivariate grey model named the GM(1,N|sin) power model is proposed. This model incorporates a power exponential term and dynamic sinusoidal function to represent the nonlinear relationship and periodic oscillations of the independent and dependent variables, respectively. Through case studies on electricity consumption and PM2.5 concentrations, the GM(1,N|sin) power model outperforms alternative models in accurately predicting time series with periodic oscillations.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Energy & Fuels
Gang Wu, Sijie Li, Shitu Abubakar, Yuqiang Li
Summary: This study simulated the composite regeneration process of a Catalyzed Diesel Particulate Filter (CDPF) and analyzed the key factors for optimizing the system.
Article
Automation & Control Systems
Xinghan Xu, Weijie Ren
Summary: This paper proposes a novel KRLS algorithm, named RFF-RMMC algorithm, which improves the prediction efficiency and robustness of KRLS algorithm by combining random Fourier feature method and maximum mixture correntropy criterion. The RFF method is used to approximate the kernel function in KRLS, reducing computational complexity and improving prediction efficiency. With flexible parameter settings, the MMCC enhances the accuracy of similarity measurement between predicted and true values.
Article
Automation & Control Systems
Chaoxu Mu, Ke Wang, Tie Qiu
Summary: This article introduces a novel event-triggering strategy integral reinforcement learning algorithm to reduce samples and transmission while ensuring learning performance. The core of the algorithm is policy iteration technique implemented by two neural networks.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Automation & Control Systems
Shoubo Feng, Min Han, Jiadong Zhang, Tie Qiu, Weijie Ren
Summary: The article proposes a novel multitask learning model for multivariate chaotic time-series prediction, which can learn both dynamic-shared and dynamic-specific patterns. By utilizing a special network structure design for dynamic analysis of multiple time series, the model can effectively disentangle complex relationships among multivariate chaotic time series.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Engineering, Electrical & Electronic
Huijuan Xia, Weijie Ren, Min Han
Summary: This study combines information theoretic learning with kernel adaptive filter to propose the generalized HQ correntropy (GHC) criterion by integrating the generalized correntropy criterion (GCC) and half-quadratic (HQ) optimization. A novel adaptive algorithm called kernel generalized half-quadratic correntropy conjugate gradient (KGHCG) algorithm is designed, which effectively enhances the robustness against non-Gaussian noise and greatly improves the convergence speed and filtering accuracy.
CIRCUITS SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Automation & Control Systems
Sheng Chen, Qihang Zhang, Xiaodong Dong, Xiaoyi Tao, Keqiu Li, Tie Qiu, Ivan Lee
Summary: Mobile edge computing (MEC) is increasingly popular due to its powerful computing capacities near end users or devices. This article proposes a framework called Sublessor to reduce wide area network (WAN) transmission costs for cooperative MEC providers. The key idea is to allow certain MEC providers to act as Internet transit brokers and resell network traffic at a reasonable price. The algorithm presented in this article finds the optimal number of brokers and reselling price, significantly reducing transmission costs by up to 35% according to experimental results.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Computer Science, Information Systems
Xiaobo Zhou, Shuxin Ge, Tie Qiu, Keqiu Li, Mohammed Atiquzzaman
Summary: Mobile edge computing is crucial for achieving ultra-low latency in 5G and beyond, by deploying services at the network edge. However, service migration in multi-user heterogeneous networks is challenging due to the difficulty in predicting user trajectories and interference among users. In this study, an optimization problem was formulated to minimize energy consumption and satisfy service latency requirements, and an efficient online algorithm called EGO was developed to solve the problem without predicting user trajectories.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Computer Science, Information Systems
Xiaoqiang Zhu, Tie Qiu, Wenyu Qu, Xiaobo Zhou, Mohammed Atiquzzaman, Dapeng Oliver Wu
Summary: This paper presents a novel indoor wireless fingerprint localization algorithm based on a broad learning system, which utilizes channel state information to overcome the problems of data loss, noise interference, and time-consuming offline training. Experimental results show that the algorithm outperforms several machine learning algorithms and existing methods in terms of training time reduction and accuracy.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Automation & Control Systems
Weijie Ren, Dewei Ma, Min Han
Summary: More and more time series data are appearing in various fields, and predicting multivariate time series is crucial to solving industrial problems. The echo state network (ESN) model has been widely used for time series prediction, but selecting suitable reservoir parameters and input feature sets remain challenging. This study proposes a modified binary salp swarm algorithm-based optimization ESN (MBSSA-ESN) model for multivariate time series prediction, which simultaneously optimizes parameter selection and feature subset. The proposed model achieves the best results compared to other methods, demonstrating its competitiveness in multivariate time series prediction.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Computer Science, Information Systems
Xiaoqiang Zhu, Tie Qiu, Wenyu Qu, Xiaobo Zhou, Yifan Wang, Dapeng Oliver Wu
Summary: In this paper, a novel data collection strategy based on reinforcement learning is proposed for fingerprint localization. By utilizing multivariate Gaussian process and mutual information, a small amount of real data is collected in advance to predict the rewards of sampling points. Then, an optimization problem is transformed into a sequential decision process and A3C algorithm is used to exploit the informative path. Extensive experiments validate the performance of the proposed algorithm and compared to existing algorithms, our system achieves similar indoor localization accuracy while reducing the workload of CSI collection.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Xiaodong Na, Mengyuan Zhang, Weijie Ren, Min Han
Summary: Multistep-ahead chaotic time series prediction is a challenging task that requires high nonlinearity and dynamical memory from the model. The proposed HESN-ARF overcomes the tradeoff between nonlinearity and memory by using hierarchical strategy and augmented random features. It achieves excellent performance in multistep-ahead chaotic time series prediction by mining and learning the latent evolution patterns in the dynamic system.
IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS
(2023)
Article
Automation & Control Systems
Min Han, Huijuan Xia, Weijie Ren
Summary: This paper proposes a dynamic model called kernel general loss algorithm based on evolving participatory learning (EPL-KGLA) for online time series prediction. The algorithm can autonomously adjust its structure and parameters to adapt to complex environments, accurately capturing the dynamic changes of time series. EPL based on evolving fuzzy systems is used in recursive clustering to utilize useful information in data streams and generate/prune structures to ensure compactness and reduce computational burden. The general loss function is combined with online kernel learning to update consequent parameters in real-time, capturing the dynamic features of data streams and improving prediction accuracy by avoiding the negative effects of anomalies or noise.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Xiaodong Na, Yuan Li, Weijie Ren, Min Han
Summary: A physics-informed hierarchical echo state network (Pi-HESN) is proposed for predicting the dynamics of chaotic systems. It captures latent evolutionary patterns hidden in the data layer by layer and integrates data and physical laws to ensure physical consistency. Experimental results demonstrate that Pi-HESN outperforms the original ESN and existing hierarchical ESN-based models in accuracy and predictability horizon on four classical chaotic systems.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Xiaodong Na, Weijie Ren, Moran Liu, Min Han
Summary: This study proposes a new method called HESN-SL for multidimensional chaotic time series prediction. The method employs stacked reservoirs to mine and capture hidden latent evolution patterns, and uses sparse learning and variable selection capability to train the output layer. Experimental results demonstrate that HESN-SL outperforms other methods in prediction performance.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Cuiying Huo, Dongxiao He, Chundong Liang, Di Jin, Tie Qiu, Lingfei Wu
Summary: In this work, we propose a new GNN-based trust evaluation method named TrustGNN, which integrates the propagative and composable nature of trust graphs into a GNN framework for better trust evaluation. TrustGNN designs specific propagative patterns for different propagative processes of trust, and distinguishes the contribution of different propagative processes to create new trust. Experiments show that TrustGNN significantly outperforms the state-of-the-art methods on widely-used real-world datasets.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Hardware & Architecture
Tie Qiu, Xinwei Yang, Ning Chen, Songwei Zhang, Geyong Min, Dapeng Oliver Wu
Summary: This study proposes a self-adaptive robustness optimization method for large-scale IoT applications, utilizing evolutionary multi-agent and distributed training mechanism to improve connectivity and lifetime. Experimental results show that this method outperforms other learning-based methods in terms of efficiency and robustness.
IEEE-ACM TRANSACTIONS ON NETWORKING
(2023)