1
|
Ye R, Zhang B, Li X, Ye Y. PEPNet: A barotropic primitive equations-based network for wind speed prediction. Neural Netw 2023; 167:533-550. [PMID: 37696071 DOI: 10.1016/j.neunet.2023.08.042] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/08/2023] [Revised: 08/02/2023] [Accepted: 08/23/2023] [Indexed: 09/13/2023]
Abstract
In wind speed prediction technologies, deep learning-based methods have achieved promising advantages. However, most existing methods focus on learning implicit knowledge in a data-driven manner but neglect some explicit knowledge from the physical theory of meteorological dynamics, failing to make stable and long-term predictions. In this paper, we explore introducing explicit physical knowledge into neural networks and propose Physical Equations Predictive Network (PEPNet) for multi-step wind speed predictions. In PEPNet, a new neural block called the Augmented Neural Barotropic Equations (ANBE) block is designed as its key component, which aims to capture the wind dynamics by combining barotropic primitive equations and deep neural networks. Specifically, the ANBE block adopts a two-branch structure to model wind dynamics, where one branch is physic-based and the other is data-driven-based. The physic-based branch constructs temporal partial derivatives of meteorological elements (including u-component wind, v-component wind, and geopotential height) in a new Neural Barotropic Equations Unit (NBEU). The NBEU is developed based on the barotropic primitive equations mode in numerical weather prediction (NWP). Besides, considering that the barotropic primitive mode is a crude assumption of atmospheric motion, another data-driven-based branch is developed in the ANBE block, which aims at capturing meteorological dynamics beyond barotropic primitive equations. Finally, the PEPNet follows a time-variant structure to enhance the model's capability to capture wind dynamics over time. To evaluate the predictive performance of PEPNet, we have conducted several experiments on two real-world datasets. Experimental results show that the proposed method outperforms the state-of-the-art techniques and achieve optimal performance.
Collapse
Affiliation(s)
- Rui Ye
- School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen 518055, Guangdong, China
| | - Baoquan Zhang
- School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen 518055, Guangdong, China.
| | - Xutao Li
- School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen 518055, Guangdong, China
| | - Yunming Ye
- School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen 518055, Guangdong, China.
| |
Collapse
|
2
|
Adli Zakaria MN, Ahmed AN, Abdul Malek M, Birima AH, Hayet Khan MM, Sherif M, Elshafie A. Exploring machine learning algorithms for accurate water level forecasting in Muda river, Malaysia. Heliyon 2023; 9:e17689. [PMID: 37456046 PMCID: PMC10344711 DOI: 10.1016/j.heliyon.2023.e17689] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/09/2023] [Revised: 06/23/2023] [Accepted: 06/26/2023] [Indexed: 07/18/2023] Open
Abstract
Accurate water level prediction for both lake and river is essential for flood warning and freshwater resource management. In this study, three machine learning algorithms: multi-layer perceptron neural network (MLP-NN), long short-term memory neural network (LSTM) and extreme gradient boosting XGBoost were applied to develop water level forecasting models in Muda River, Malaysia. The models were developed using limited amount of daily water level and meteorological data from 2016 to 2018. Different input scenarios were tested to investigate the performance of the models. The results of the evaluation showed that the MLP model outperformed both the LSTM and the XGBoost models in predicting water levels, with an overall accuracy score of 0.871 compared to 0.865 for LSTM and 0.831 for XGBoost. No noticeable improvement has been achieved after incorporating meteorological data into the models. Even though the lowest reported performance was reported by the XGBoost, it is the faster of the three algorithms due to its advanced parallel processing capabilities and distributed computing architecture. In terms of different time horizons, the LSTM model was found to be more accurate than the MLP and XGBoost model when predicting 7 days ahead, demonstrating its superiority in capturing long-term dependencies. Therefore, it can be concluded that each ML model has its own merits and weaknesses, and the performance of different ML models differs on each case because these models depends largely on the quantity and quality of data available for the model training.
Collapse
Affiliation(s)
- Muhamad Nur Adli Zakaria
- Department of Civil Engineering, College of Engineering, Universiti Tenaga Nasional (UNITEN), 43000, Kajang, Selangor, Malaysia
| | - Ali Najah Ahmed
- Department of Civil Engineering, College of Engineering, Universiti Tenaga Nasional (UNITEN), 43000, Kajang, Selangor, Malaysia
- Institute of Energy Infrastructure (IEI) , Universiti Tenaga Nasional (UNITEN), 43000, Selangor, Malaysia
| | - Marlinda Abdul Malek
- Cataclysmic Management and Sustainable Development Research Group (CAMSDE), Department of Civil Engineering, Kulliyyah of Engineering, International Islamic University Malaysia, Malaysia
| | - Ahmed H. Birima
- Department of Civil Engineering, College of Engineering, Qassim University, Unaizah, Saudi Arabia
| | - Md Munir Hayet Khan
- Faculty of Engineering & Quantity Surveying, INTI International University (INTI-IU), Persiaran Perdana BBN, Putra Nilai, Nilai, 71800, Negeri Sembilan, Malaysia
| | - Mohsen Sherif
- Civil and Environmental Eng. Dept., College of Engineering, United Arab Emirates University, Al Ain, 15551, United Arab Emirates
- National Water and Energy Center, United Arab Emirates University, P.O. Box. 15551, Al Ain, United Arab Emirates
| | - Ahmed Elshafie
- Department of Civil Engineering, Faculty of Engineering, University of Malaya (UM), 50603, Kuala Lumpur, Malaysia
| |
Collapse
|
3
|
Xu Y, Wang C, Liang J, Yue K, Li W, Zheng S, Zhao Z. Deep Reinforcement Learning Based Decision Making for Complex Jamming Waveforms. ENTROPY 2022; 24:1441. [PMCID: PMC9601320 DOI: 10.3390/e24101441] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 09/17/2022] [Accepted: 10/07/2022] [Indexed: 06/18/2023]
Abstract
With the development of artificial intelligence, intelligent communication jamming decision making is an important research direction of cognitive electronic warfare. In this paper, we consider a complex intelligent jamming decision scenario in which both communication parties choose to adjust physical layer parameters to avoid jamming in a non-cooperative scenario and the jammer achieves accurate jamming by interacting with the environment. However, when the situation becomes complex and large in number, traditional reinforcement learning suffers from the problems of failure to converge and a high number of interactions, which are fatal and unrealistic in a real warfare environment. To solve this problem, we propose a deep reinforcement learning based and maximum-entropy-based soft actor-critic (SAC) algorithm. In the proposed algorithm, we add an improved Wolpertinger architecture to the original SAC algorithm in order to reduce the number of interactions and improve the accuracy of the algorithm. The results show that the proposed algorithm shows excellent performance in various scenarios of jamming and achieves accurate, fast, and continuous jamming for both sides of the communication.
Collapse
Affiliation(s)
- Yuting Xu
- Key Laboratory of RF Circuits and Systems, Ministry of Education, Hangzhou Dianzi University, Hangzhou 310018, China
| | - Chao Wang
- Key Laboratory of RF Circuits and Systems, Ministry of Education, Hangzhou Dianzi University, Hangzhou 310018, China
| | - Jiakai Liang
- Key Laboratory of RF Circuits and Systems, Ministry of Education, Hangzhou Dianzi University, Hangzhou 310018, China
| | - Keqiang Yue
- Key Laboratory of RF Circuits and Systems, Ministry of Education, Hangzhou Dianzi University, Hangzhou 310018, China
- Science and Technology on Communication Information Security Control Laboratory, The No. 011 Research Center, Jiaxing 314033, China
| | - Wenjun Li
- Key Laboratory of RF Circuits and Systems, Ministry of Education, Hangzhou Dianzi University, Hangzhou 310018, China
| | - Shilian Zheng
- Science and Technology on Communication Information Security Control Laboratory, The No. 011 Research Center, Jiaxing 314033, China
| | - Zhijin Zhao
- The School of Communication Engineering, Hangzhou Dianzi University, Hangzhou 310018, China
| |
Collapse
|
4
|
Zamri NE, Azhar SA, Mansor MA, Alway A, Kasihmuddin MSM. Weighted Random k Satisfiability for k=1,2 (r2SAT) in Discrete Hopfield Neural Network. Appl Soft Comput 2022. [DOI: 10.1016/j.asoc.2022.109312] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/02/2022]
|