Article
Engineering, Civil
Shu Chen, Miaomiao Ren, Wei Sun
Summary: By combining a hybrid model based on two-stage decomposition, support vector machine, and a combined method, the accuracy of annual runoff forecasting is significantly improved. An empirical study conducted at the Pingshi Station in the Lechangxia Basin demonstrates that the two-stage decomposition greatly enhances forecasting ability, with the combined model outperforming the member models.
JOURNAL OF HYDROLOGY
(2021)
Article
Engineering, Civil
Xuehua Zhao, Hanfang Lv, Shujin Lv, Yuting Sang, Yizhao Wei, Xueping Zhu
Summary: This study proposes a new hydrological prediction model ICEEWT-IGWO-GRU, which combines empirical wavelet transform and improved complete ensemble empirical mode decomposition with adaptive noise, applies gated recurrent unit deep learning, and improved grey wolf optimizer to enhance accuracy and robustness of streamflow prediction. In comparison with other models, the results show that this model demonstrates superior performance in monthly streamflow forecasting.
JOURNAL OF HYDROLOGY
(2021)
Article
Engineering, Multidisciplinary
Guohui Li, Zelin Yang, Hong Yang
Summary: This study aims to construct a decomposition-prediction model to predict carbon emissions in aviation using optimized methods. The results show that the proposed method has the best prediction performance for domestic and international aviation.
ALEXANDRIA ENGINEERING JOURNAL
(2023)
Article
Automation & Control Systems
Zicheng Wang, Liren Chen, Huayou Chen, Naveed ur Rehman
Summary: This paper proposes a novel multiscale and multivariable methodology based on multivariate variational mode decomposition (MVMD) and machine learning (ML) algorithms for ship price forecasting. Empirical results show that the proposed methodology outperforms benchmark models in terms of both level and directional forecasting accuracy, indicating its potential value for shipping market analysis and forecasting.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Zichen Zhang, Wei -Chiang Hong
Summary: Accurate electric load forecasting is crucial for the efficiency of power system operation. Hybrid intelligent computing methods and swarm-based algorithms, along with the SVR model, show promising results in solving convergence issues. The proposed VMD-SVR-CGWO model outperforms other models in forecasting accuracy based on numerical examples from two electric load data sets.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Tufan Inac, Emrah Dokur, Ugur Yuzgec
Summary: The Multi-strategy Random weighted Gray Wolf Optimizer (MsRwGWO) is proposed as an enhancement of the original GWO algorithm, with new mechanisms added to improve search performance. Evaluation on CEC 2014 test suite and analysis of algorithm behavior during optimization process were conducted to verify performance improvements.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Engineering, Marine
Xiang Ji, Zhe Tian, Hong Song, Fushun Liu
Summary: This paper presents a method for assessing the performance degradation of offshore wind turbine structures based on an optimized variational mode decomposition algorithm. The method accurately identifies the reduction in structural stiffness and demonstrates its effectiveness through numerical calculations, physical tests, and field measurements.
Article
Green & Sustainable Science & Technology
Chenhui Wang, Wei Guo
Summary: Accurate prediction of landslide displacement is achieved by establishing a prediction model based on VMD and GWO-SVR. The original data are decomposed using VMD, and influential factors are selected for constructing the input training data set. The model achieves good prediction accuracy by summing up the three displacement components.
Article
Engineering, Civil
Tonglin Fu, Xinrong Li, Rongliang Jia, Li Feng
Summary: The study proposed a novel hybrid estimation approach to estimate monthly evapotranspiration (ET) using historical ET time series in the southeastern margins of the Tengger Desert, China. The hybrid model based on variational mode decomposition (VMD), grey wolf optimizer (GWO), and support vector machine (SVM) achieved superior computational performance compared to other methods, with an increase in Nash-Sutcliffe coefficient of efficiency (NSCE) and decrease in mean absolute percentage error (MAPE) during testing. The hybrid VMD-GWO-SVM model is suggested as the best choice for estimating ET in the absence of regional meteorological monitoring.
JOURNAL OF HYDROLOGY
(2021)
Article
Energy & Fuels
Mengran Zhou, Tianyu Hu, Kai Bian, Wenhao Lai, Feng Hu, Oumaima Hamrani, Ziwei Zhu
Summary: Short-term electric load forecasting is crucial for the safe and stable operation of power systems and transactions in the power market. This paper proposes an approach using decomposition and ensemble framework with verification on load data from Oslo and surrounding areas in Norway, achieving optimal evaluation metrics for short-term electric load forecasting and showing good application prospects.
Article
Energy & Fuels
Yumeng Huang, Xingyu Dai, Qunwei Wang, Dequn Zhou
Summary: The study introduces a novel carbon price forecasting model, utilizing the VMD-GARCH/LSTM-LSTM model with a decomposition-ensemble approach, combining econometric and artificial intelligence techniques to accurately predict the rapidly rising and volatile carbon prices in the current market context.
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
Computer Science, Artificial Intelligence
Aytac Altan, Seckin Karasu, Enrico Zio
Summary: A new hybrid WSF model is developed in this paper, combining LSTM and decomposition methods, optimized with GWO. The experimental results show that the proposed hybrid model can capture the non-linear characteristics of WSTS, achieving better forecasting accuracy than single forecasting models.
APPLIED SOFT COMPUTING
(2021)
Article
Engineering, Civil
Junhao Wu, Zhaocai Wang, Yuan Hu, Sen Tao, Jinghan Dong
Summary: Water resources play a significant role in sustaining the biological and socio-economic development of a region. Due to factors like basin geography and climate change, runoff variability exhibits non-linear and non-stationary characteristics. Runoff forecasting is crucial for preventing flood disasters, but improving its accuracy remains a challenge in water resources management. In this study, an ensemble deep learning model was developed to forecast daily river runoff, using variational mode decomposition (VMD) for feature extraction and a bi-directional long short-term memory network (BiLSTM) with an attention mechanism (AM) for prediction. The model showed better performance compared to other models, indicating its potential as an effective tool for hydrological forecasting and water resources management.
WATER RESOURCES MANAGEMENT
(2023)
Article
Engineering, Multidisciplinary
Xiaoan Yan, Wangji Yan, Ka-Veng Yuen, Zhixin Yang, Xianbo Wang
Summary: This paper proposes an adaptive variational mode extraction method based on multi-domain and multi-objective optimization (AVME-MDMO) to improve bearing fault feature extraction and diagnosis. The method automatically determines key parameters using a novel multi-objective function and demonstrates superiority over existing methods in excavating fault signatures.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2023)
Article
Computer Science, Interdisciplinary Applications
Wen-chuan Wang, Lei Xu, Kwok-wing Chau, Yong Zhao, Dong-mei Xu
Summary: Yin-Yang-pair Optimization (YYPO) is a philosophy-inspired meta-heuristic algorithm that generates candidate solutions by balancing exploitation and exploration, but suffers from low solution quality in exploration. To enhance performance, a new algorithm named orthogonal opposition-based-learning Yin-Yang-pair Optimization (OOYO) is proposed, which utilizes orthogonal experiment design and opposition-based learning to optimize candidate solutions.
ENGINEERING WITH COMPUTERS
(2022)
Article
Computer Science, Artificial Intelligence
Wen-chuan Wang, Lei Xu, Kwok-wing Chau, Chang-jun Liu, Qiang Ma, Dong-mei Xu
Summary: This paper proposes a lightweight and efficient variant of differential evolution algorithm, Ce-LDE, for solving constrained single-objective optimization problems. The algorithm achieves high competitiveness and practicality through the introduction of a combined constraint handling method and redefinition of control parameters, as demonstrated by experimental results and comparative studies.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Environmental Sciences
Wenchuan Wang, Yanwei Zhao, Yong Tu, Rui Dong, Qiang Ma, Changjun Liu
Summary: This study examines the influence of different methods on the parameter regionalization of distributed hydrological model parameters in hilly areas of Hunan Province. Six parameter regionalization schemes are proposed, and 426 flood events are used for model parameter calibration, with 136 events for validation. The results show that the random forest model is the most stable solution and significantly outperforms other methods. Using the random forest model for parameter regionalization can improve the accuracy of flood simulation in ungauged areas, which is of great significance for flash flood forecasting and early warning.
Article
Environmental Sciences
Sijia Hao, Wenchuan Wang, Qiang Ma, Changzhi Li, Lei Wen, Jiyang Tian, Changjun Liu
Summary: With limited data, disaster simulation review based on digital information technology is an important guideline for analyzing disaster mechanisms, planning post-disaster reconstruction, and improving defense capability. By using limited measured data, a hydrological-hydrodynamic model was established to study the 7.20 flash flood in the Wangzongdian river basin. The results showed that extreme rainstorm caused flooding in mountainous areas and the collapse of subgrade water damming, contributing to the serious disaster.
Article
Engineering, Environmental
Mingwei Ma, Zhaohang Wang, Huijuan Cui, Wenchuan Wang, Liuyuwei Jiang
Summary: This study constructs a multi-scalar framework for attribution analysis by integrating hydrological modeling into the Budyko-based decomposition method and applies it to the source region of the Yellow River as a case study. The results indicate that climate change is the dominant factor controlling streamflow variation for the annual and wet season, while human activities play a major role in streamflow variation for the dry season. The applicability of the Budyko-based decomposition method within the new multi-scalar framework is verified through hydrologic simulation.
Article
Computer Science, Interdisciplinary Applications
Dong-mei Xu, Xiang Wang, Wen-chuan Wang, Kwok-wing Chau, Hong-fei Zang
Summary: In this study, a coupled forecasting model combining ICEEMDAN, WD, and SVM optimized by SOA is proposed to predict monthly runoff. The model decomposes the original runoff series using ICEEMDAN and WD to obtain IMF and Res components, which are then input into the SOA-SVM model for prediction. The ICEEMDAN-WD-SOA-SVM model achieves the smallest RMSE and MAPE and the largest NSEC and R compared to other benchmarking models, demonstrating its superior prediction accuracy.
JOURNAL OF HYDROINFORMATICS
(2023)
Article
Engineering, Civil
Wen-chuan Wang, Qi Cheng, Kwok-wing Chau, Hao Hu, Hong-fei Zang, Dong-mei Xu
Summary: Reliable runoff prediction is essential for reservoir scheduling, water resources management, and efficient water utilization. To improve the accuracy of monthly runoff prediction, a hybrid model (TVF-EMD-SSA-ELM) combining TVF-based EMD, SSA, and ELM is proposed. The model successfully decomposes the runoff series, optimizes the ELM model with SSA, and generates accurate predictions. Evaluation results show that the TVF-EMD-SSA-ELM model outperforms other models in terms of prediction accuracy. This model provides a new method for monthly runoff prediction and can be applied in similar study areas.
JOURNAL OF HYDROLOGY
(2023)
Review
Engineering, Civil
Wen-chuan Wang, Wei-can Tian, Dong-mei Xu, Kwok-wing Chau, Qiang Ma, Chang-jun Liu
Summary: River flood routing is a crucial aspect of water resources management, with the Muskingum model being the dominant method. This paper reviews the development and parameter estimation research status of the Muskingum model. The combination of mathematical techniques and evolutionary algorithms has shown promising results in recent years. The paper also provides an overview of accuracy evaluation criteria and research case data sets commonly used in the literature, and discusses challenges and future trends in Muskingum model research.
WATER RESOURCES MANAGEMENT
(2023)
Article
Computer Science, Interdisciplinary Applications
Dong-mei Xu, Xiao-xue Hu, Wen-chuan Wang, Kwok-wing Chau, Hong-fei Zang
Summary: This research provides a hybrid forecasting model to increase the precision of monthly runoff predictions. It applies complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and variational mode decomposition (VMD) to decompose the raw monthly runoff time series. The input-output relationships for all intrinsic mode functions (IMFs) are determined using Harris Hawks Optimization (HHO) algorithm to optimize least squares support vector machine (LSSVM) model. Evaluation indicators demonstrate the effectiveness of the proposed hybrid model in improving prediction accuracy.
EARTH SCIENCE INFORMATICS
(2023)
Article
Environmental Sciences
Wenchuan Wang, Weican Tian, Kwokwing Chau, Hongfei Zang, Mingwei Ma, Zhongkai Feng, Dongmei Xu
Summary: This paper proposes an improved bald eagle search algorithm (CABES) combined with epsilon-constraint method (epsilon-CABES) to tackle the complex reservoir flood control operation problem. Through simulations and comparisons with other algorithms, the superior performance of the CABES algorithm is verified. The results of the tests on single and multi-reservoir systems show that the epsilon-CABES method outperforms other methods in flood control scheduling.
Article
Computer Science, Interdisciplinary Applications
Yi-yang Wang, Wenchuan Wang, Kwok-wing Chau, Dong-mei Xu, Hong-fei Zang, Chang-jun Liu, Qiang Ma
Summary: This article proposes a multi-head attention flood forecasting model (MHAFFM) that combines a multi-head attention mechanism with multiple linear regression for precise and stable multi-hour flood forecasting. Experimental results show that the MHAFFM model significantly improves the prediction performance compared to benchmarking models, while maintaining good stability and interpretability. This research enhances the credibility of deep learning in the field of hydrology and provides a new approach for its application.
JOURNAL OF HYDROINFORMATICS
(2023)
Article
Engineering, Civil
Zhong-kai Feng, Wen-jing Niu, Tai-heng Zhang, Wen-chuan Wang, Tao Yang
Summary: To address the practical requirement, this research proposes a novel artificial intelligence method for deriving reservoir operation policy, which uses fuzzy clustering iteration method and novel twin support vector regression model. The feasibility of the proposed method is evaluated on two real-world huge hydropower reservoirs in China, and the simulations demonstrate better comprehensive benefits than several control methods under uncertain environments. Therefore, the experiments confirm that metaheuristic algorithms and pattern recognition techniques can enhance the performance of standalone artificial intelligence methods in deriving reservoir operation policy.
JOURNAL OF HYDROLOGY
(2023)
Article
Environmental Sciences
Mingwei Ma, Hongfei Zang, Wenchuan Wang, Huijuan Cui, Yanwei Sun, Yujia Cheng
Summary: In this study, a copula-based approach was used to propose the classical severity-duration-frequency (SDF) relationships of streamflow drought in the source area of the Yellow River. Multiple time-varying threshold levels and the integration and elimination of drought events were considered. The findings show that the copula-based SDF relationships can provide more critical information than univariate frequency analysis, as they effectively consider the connection and interaction between drought characteristics.
Article
Engineering, Multidisciplinary
Wenchuan Wang, Weican Tian, Kwok-wing Chau, Yiming Xue, Lei Xu, Hongfei Zang
Summary: The improved Bald Eagle algorithm (CABES) enhances the performance of the Bald Eagle Search algorithm (BES) by integrating Cauchy mutation and adaptive optimization. CABES adjusts the step size in the selection stage to select a better search range, and updates the search position formula with an adaptive weight factor to further improve the local optimization capability of BES. Experimental results demonstrate that CABES exhibits good exploration and exploitation abilities, making it effective and efficient in practical engineering problems.
CMES-COMPUTER MODELING IN ENGINEERING & SCIENCES
(2023)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Wen-Chuan Wang, Yan-Wei Zhao, Chang-Jun Liu, Qiang Ma, Dong-Mei Xu
Summary: This paper uses machine learning methods to improve the accuracy of flood simulation and early warning for ungauged areas, with a case study in Hunan Province. The results show that the regionalization scheme based on the random forest model significantly improves flood simulation accuracy, and the support vector machine-based warning model continues to improve and is expected to reach a high level of accuracy in the coming years.
ADVANCES IN HYDROINFORMATICS: MODELS FOR COMPLEX AND GLOBAL WATER ISSUES-PRACTICES AND EXPECTATIONS
(2022)