Article
Ecology
Jun Zhang, Xuan Zhang
Summary: In this paper, a carbon neutral energy saving and emission reduction prediction model for sports competitions is proposed, which is based on an attention mechanism-based convolutional neural network (CNN) combined with the gated recurrent unit (GRU). The model collects real-time carbon emissions data from sports events and uses deep learning algorithms to predict and compare carbon emissions from sports competitions. This model has a certain reference value in identifying energy saving and emission reduction measures for carbon neutral sports events.
FRONTIERS IN ECOLOGY AND EVOLUTION
(2023)
Article
Energy & Fuels
Zulfiqar Ahmad Khan, Tanveer Hussain, Sung Wook Baik
Summary: In this research, a dual-stream network is proposed for accurate photovoltaic forecasting. It extracts features by parallel learning of spatial patterns and sequential learning algorithm, and selects optimal features for forecasting using self-attention mechanism. The network is derived after a series of experiments, and achieves higher forecasting accuracy compared to state-of-the-art models.
Article
Agriculture, Multidisciplinary
Yue Gao, Kai Yan, Baisheng Dai, Hongmin Sun, Yanling Yin, Runze Liu, Weizheng Shen
Summary: This study proposes a hybrid model that combines convolutional neural network (CNN) and gated recurrent unit (GRU) to differentiate aggressive and other behaviors from surveillance videos. The proposed model outperforms the state-of-the-art approaches in aggressive behavior recognition by integrating CNN, GRU, and a specific spatio-temporal attention mechanism.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2023)
Article
Mathematics
Harshwardhan Yadav, Param Shah, Neel Gandhi, Tarjni Vyas, Anuja Nair, Shivani Desai, Lata Gohil, Sudeep Tanwar, Ravi Sharma, Verdes Marina, Maria Simona Raboaca
Summary: Cardiovascular diseases (CVDs) are a significant cause of death worldwide. Deep learning algorithms have been utilized to detect various types of heartbeat sounds at an early stage, improving the accuracy of classification. The proposed CNN + BiGRU attention-based architecture achieves an accuracy of 90% in heartbeat sound classification, outperforming other DL models.
Article
Engineering, Civil
Peng Mei, Meng Li, Qian Zhang, Ginlin Li, Lang Song
Summary: Accurately predicting the quality of raw water is of great significance for the operation and management of waterworks. In this study, a CGA model was proposed to predict the turbidity and CODcr of raw water with light industrial-agricultural pollution, based on long-term water quality data. The CGA model, which combines CNN, GRU, and attention layers, outperforms individual LSTM and GRU models. The study findings also indicate a gradual decline in the overall raw water quality due to local industrial-agricultural pollution.
JOURNAL OF HYDROLOGY
(2022)
Article
Green & Sustainable Science & Technology
Jingming Su, Xuguang Han, Yan Hong
Summary: A new algorithm, PSVMD-CGA, is proposed in this paper to optimize variational mode decomposition (VMD) by combining CNN, GRU, and an attention mechanism with the Sparrow algorithm. The PSVMD-CGA model has been significantly enhanced compared to the original GRU model, with a decrease in MAE by 288.8%, MAPE by 3.46%, RMSE by 326.1 MW, and an increase in R2 to 0.99. Various evaluation indicators show that the PSVMD-CGA model outperforms the SSA-VMD-CGA and GA-VMD-CGA models.
Article
Computer Science, Artificial Intelligence
Lang He, Jonathan Cheung-Wai Chan, Zhongmin Wang
Summary: The study proposes an integrated framework DLGA-CNN for depression recognition, combining CNN with attention mechanism and weighted spatial pyramid pooling. By focusing on local and global attention, it effectively mines depression patterns from facial videos.
Article
Green & Sustainable Science & Technology
Dinggao Liu, Zhenpeng Tang, Yi Cai
Summary: This study develops a hybrid prediction model to accurately predict the volatile soybean spot price in China. By combining component clustering and attention mechanism with a neural network, the model's prediction ability is improved. Empirical analysis shows that the proposed model outperforms other techniques and algorithms in terms of prediction accuracy.
Article
Computer Science, Artificial Intelligence
Wanzhi Wen, Tian Zhao, Shiqiang Wang, Jiawei Chu, Deepak Kumar Jain
Summary: This paper proposes a code recommendation method based on joint embedded attention network (JEAN) to address the heterogeneity of program language and natural language query. By using GRU Network for embedding and describing, the method solves the problem of heterogeneous code snippets and queries. The Attention mechanism is used to distribute different weights to different components, making the code recommendation more interpretable. Experimental results demonstrate that the proposed method outperforms other baseline models in recommending appropriate code snippets for developers' needs.
Article
Energy & Fuels
Peiwen Yang, Debin Fang, Shuyi Wang
Summary: This study proposes a distributed power agency trading mechanism based on dynamic game theory and deviation assessment for local energy markets with integrated distributed power prosumers. Simulation results show that prosumers prefer this mechanism during peak load hours at noon and it provides higher carbon reduction compared to other mechanisms.
Article
Chemistry, Multidisciplinary
Kun Wang, Yuan Tan, Lizhong Zhang, Zhigang Chen, Jinghong Lei
Summary: Fault early warning is a challenge in operation and maintenance, and this research proposes a targeted solution by combining spatiotemporal characteristics and topological characteristics in the TCAG model to improve fault prediction accuracy.
APPLIED SCIENCES-BASEL
(2022)
Article
Engineering, Civil
Yangying He, Linying Chen, Junmin Mou, Qingsong Zeng, Yamin Huang, Pengfei Chen, Song Zhang
Summary: This paper introduces a novel bio-inspired method that utilizes the favorable hydrodynamic interaction in ship convoys to achieve an energy bonus for the whole formation system or members. Through simulations and experiments, the feasibility of this method is verified, and a model is established to evaluate the resistance characteristics of ships in different configurations and velocities, finding the optimal configuration. Computational fluid dynamics results show that sailing in formation can achieve fuel savings and emission reduction.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Environmental Sciences
Anping Wan, Jie Yang, Ting Chen, Yang Jinxing, Ke Li, Zhou Qinglong
Summary: This paper proposes a pollution emission prediction method for CHP systems based on feature engineering and a hybrid deep learning model. The effectiveness of the method is verified through a case study using actual dataset.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2022)
Article
Engineering, Mechanical
Bin Wang, Yanbao Guo, Deguo Wang, Yuansheng Zhang, Renyang He, Jinzhong Chen
Summary: This paper proposes a time series prediction method based on acoustic emission signals, which can effectively predict crack evolution in natural gas pipelines. It improves the prediction accuracy and is significant for the safe operation of gas projects.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Engineering, Industrial
Yuanfu Li, Yifan Chen, Haonan Shao, Huisheng Zhang
Summary: Improving Remaining Useful Life (RUL) prediction accuracy is the main focus in Prognostic and Health Management (PHM). Deep learning offers opportunities to address reliability engineering and safety analysis challenges, but performance can be enhanced with better sensor data quality and quantity. This paper proposes a novel RUL prediction approach that combines an encoder/decoder architecture with Gated Recurrent Units (GRUs) and a dual attention mechanism, leveraging domain knowledge. Compared to other attention-based deep learning models, the approach achieved first place in overall ranking on the NASA C-MAPSS dataset, demonstrating the reliability of incorporating domain knowledge in deep learning training. The results indicate superior estimation accuracy, efficiency, and interpretability of the proposed approach.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)