Siyuan He
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Siyuan He has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Yuhao Liu
Guangdong University of Science and Technology, Dongguan, China
4 shared publications - Chengjun Chen
Guangdong University of Science and Technology, Dongguan, China
3 shared publications - Dean Xu
School of Business, Guangxi Minzu Normal University, Chongzuo, Guangxi, China
1 shared publication - Guanwei Jang
Shaoguan University Shaoguan China
1 shared publication Jin Peng
Management School, Guangdong University of Science and Technology, Dongguan, Guangdong, China
1 shared publication- WeiGuo Liang
Guangdong University of Science and Technology, Dongguan, China
1 shared publication
Latest Publications
- Open Access
- 10 September 2025
- Published by Frontiers Media SA in Frontiers in Environmental Science
- Vol. 13, 1653446
A CEEMDAN-GNN-transformer hybrid model for air quality index forecasting: a case study of Chang’an town, Dongguan, China
Air pollution has emerged as a pressing global environmental issue, and accurate forecasting plays a critical role in environmental governance and public health protection. This study proposes an enhanced air quality forecasting model based on a hybrid CEEMDAN-GNN-Transformer architecture, and conducts an empirical analysis using data from Chang’an Town, Dongguan, China. The proposed model first employs Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) to extract multi-scale temporal features and mitigate non-stationary noise in the data. Then, a Graph Neural Network (GNN) is applied to capture the spatial dependencies among various air pollutants. Finally, a Transformer model is utilized to model complex temporal dependencies and improve the capture of long-term trends. The research uses historical air quality monitoring data from 2015 to 2024, including concentrations of PM2.5, PM10, SO2, CO, NO2, and O3 as input features, with the Air Quality Index (AQI) as the prediction target. Model performance is enhanced through ablation studies and hyperparameter tuning, and is compared against several mainstream baseline models. Experimental results demonstrate that the proposed CEEMDAN-GNN-Transformer model outperforms traditional approaches in terms of MAE, MSE, and R2 metrics, achieving superior prediction accuracy and robustness. This study not only contributes to the theoretical advancement of air quality forecasting methodologies but also provides a more precise predictive tool for environmental management and public health risk prevention, offering significant practical value.
- 10 September 2025
- Published by Frontiers Media SA in Frontiers in Environmental Science
- Vol. 13, 1653446
Open AccessA CEEMDAN-GNN-transformer hybrid model for air quality index forecasting: a case study of Chang’an town, Dongguan, China
Air pollution has emerged as a pressing global environmental issue, and accurate forecasting plays a critical role in environmental governance and public health protection. This study proposes an enhanced air quality forecasting model based on a hybrid CEEMDAN-GNN-Transformer architecture, and conducts an empirical analysis using data from Chang’an Town, Dongguan, China. The proposed model first employs Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) to extract multi-scale temporal features and mitigate non-stationary noise in the data. Then, a Graph Neural Network (GNN) is applied to capture the spatial dependencies among various air pollutants. Finally, a Transformer model is utilized to model complex temporal dependencies and improve the capture of long-term trends. The research uses historical air quality monitoring data from 2015 to 2024, including concentrations of PM2.5, PM10, SO2, CO, NO2, and O3 as input features, with the Air Quality Index (AQI) as the prediction target. Model performance is enhanced through ablation studies and hyperparameter tuning, and is compared against several mainstream baseline models. Experimental results demonstrate that the proposed CEEMDAN-GNN-Transformer model outperforms traditional approaches in terms of MAE, MSE, and R2 metrics, achieving superior prediction accuracy and robustness. This study not only contributes to the theoretical advancement of air quality forecasting methodologies but also provides a more precise predictive tool for environmental management and public health risk prevention, offering significant practical value.
- Open Access
- 7 May 2025
- Published by HARD Publishing Company in Polish Journal of Environmental Studies
- Vol. 35 (3) , 3187-3207
Hybrid Transformer-TimesNet Model for Accurate Prediction of Industrial Air Pollutants: A Case Study of the Xinyang Industrial Zone, China
Air pollution in industrial zones poses significant threats to environmental and public health, especially in developing regions, highlighting the necessity for accurate forecasting to guide pollution management. This study presents an Optuna-enhanced hybrid Transformer-TimesNet model aimed at improving time series forecasting of six critical pollutants (SO2, NO2, CO, O3, PM10, and PM2.5) in Xinyang Industrial Zone, Xiamen, China. Utilizing air quality data from 2019 to 2023, the model combines the Transformer’s strength in capturing long-range dependencies with TimesNet’s expertise in handling complex temporal patterns. Advanced preprocessing techniques were employed to address both linear and non-linear data components, and Optuna was used for hyperparameter tuning, enhancing model stability and predictive accuracy. Comparative experiments demonstrated the hybrid model’s superior performance against traditional statistical methods, machine learning models, and deep learning approaches, evaluated through metrics such as MAE, RMSE, SMAPE, and R2. The model’s capability to accurately capture long-term pollutant trends underscores its reliability and validity as a predictive tool for policymakers and environmental managers. These results contribute to theoretical advancements in environmental monitoring and offer practical solutions for public health protection and pollution mitigation, demonstrating the potential of hybrid deep learning models in addressing complex forecasting challenges.
- 7 May 2025
- Published by HARD Publishing Company in Polish Journal of Environmental Studies
- Vol. 35 (3) , 3187-3207
Open AccessHybrid Transformer-TimesNet Model for Accurate Prediction of Industrial Air Pollutants: A Case Study of the Xinyang Industrial Zone, China
Air pollution in industrial zones poses significant threats to environmental and public health, especially in developing regions, highlighting the necessity for accurate forecasting to guide pollution management. This study presents an Optuna-enhanced hybrid Transformer-TimesNet model aimed at improving time series forecasting of six critical pollutants (SO2, NO2, CO, O3, PM10, and PM2.5) in Xinyang Industrial Zone, Xiamen, China. Utilizing air quality data from 2019 to 2023, the model combines the Transformer’s strength in capturing long-range dependencies with TimesNet’s expertise in handling complex temporal patterns. Advanced preprocessing techniques were employed to address both linear and non-linear data components, and Optuna was used for hyperparameter tuning, enhancing model stability and predictive accuracy. Comparative experiments demonstrated the hybrid model’s superior performance against traditional statistical methods, machine learning models, and deep learning approaches, evaluated through metrics such as MAE, RMSE, SMAPE, and R2. The model’s capability to accurately capture long-term pollutant trends underscores its reliability and validity as a predictive tool for policymakers and environmental managers. These results contribute to theoretical advancements in environmental monitoring and offer practical solutions for public health protection and pollution mitigation, demonstrating the potential of hybrid deep learning models in addressing complex forecasting challenges.
- Open Access
- 20 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Analyzing Factors Influencing the Effectiveness of Technological Transformation in Logistics Enterprises: A Study Based on the VAR Model
- 20 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessAnalyzing Factors Influencing the Effectiveness of Technological Transformation in Logistics Enterprises: A Study Based on the VAR Model
- Open Access
- 15 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 68-74
Forecasting Sales of China FAW New Energy Vehicles Using Ensemble Neural Networks
In recent years, the rapid development of the new energy vehicle (NEV) industry has been driven by the global shift in energy structures and increased environmental awareness. Accurate sales forecasting has become crucial in the automotive industry to enhance production efficiency and market responsiveness. This paper focuses on China FAW's NEVs, utilizing data from 2019 to 2023 to predict sales using an ensemble of neural network algorithms. We constructed three models—Convolutional Neural Network (CNN), Feedforward Neural Network (FNN), and Multilayer Perceptron (MLP)—and analyzed multiple factors, including automotive indices, new energy indices, and loan interest rates, to forecast sales for 2024. The experimental results indicate that CNN outperforms traditional statistical methods and other neural network models across various metrics. Additionally, integrating the three neural network models further enhances predictive performance, demonstrating high accuracy and stability. In the future, optimizing and combining multiple models may provide even more precise forecasts in complex market environments.
- 15 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 68-74
Open AccessForecasting Sales of China FAW New Energy Vehicles Using Ensemble Neural Networks
In recent years, the rapid development of the new energy vehicle (NEV) industry has been driven by the global shift in energy structures and increased environmental awareness. Accurate sales forecasting has become crucial in the automotive industry to enhance production efficiency and market responsiveness. This paper focuses on China FAW's NEVs, utilizing data from 2019 to 2023 to predict sales using an ensemble of neural network algorithms. We constructed three models—Convolutional Neural Network (CNN), Feedforward Neural Network (FNN), and Multilayer Perceptron (MLP)—and analyzed multiple factors, including automotive indices, new energy indices, and loan interest rates, to forecast sales for 2024. The experimental results indicate that CNN outperforms traditional statistical methods and other neural network models across various metrics. Additionally, integrating the three neural network models further enhances predictive performance, demonstrating high accuracy and stability. In the future, optimizing and combining multiple models may provide even more precise forecasts in complex market environments.
- Open Access
- 19 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 292-298
POY Price Prediction Model based on XGBoost Algorithm
Due to the oligopoly structure of upstream suppliers in the polyester filament yarn (POY) industry, downstream enterprises often have relatively limited bargaining power. Consequently, the continuous fluctuations in POY prices significantly impact the relevant costs of downstream production enterprises. Accurately predicting POY price changes is crucial for these enterprises to formulate optimized procurement strategies, effectively reducing overall costs and promoting stable and healthy development. This study utilizes machine learning algorithms to analyze POY price forecasting based on data from 2014 to 2023, aiming to provide reference for enterprises in formulating optimal material procurement strategies. Analysis reveals that the XGBoost model demonstrates high accuracy and predictive capability in forecasting POY prices. Based on the model's predictions, we derive the trend of POY prices for 2024, thereby providing a reference for enterprises to formulate optimal material procurement strategies.
- 19 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 292-298
Open AccessPOY Price Prediction Model based on XGBoost Algorithm
Due to the oligopoly structure of upstream suppliers in the polyester filament yarn (POY) industry, downstream enterprises often have relatively limited bargaining power. Consequently, the continuous fluctuations in POY prices significantly impact the relevant costs of downstream production enterprises. Accurately predicting POY price changes is crucial for these enterprises to formulate optimized procurement strategies, effectively reducing overall costs and promoting stable and healthy development. This study utilizes machine learning algorithms to analyze POY price forecasting based on data from 2014 to 2023, aiming to provide reference for enterprises in formulating optimal material procurement strategies. Analysis reveals that the XGBoost model demonstrates high accuracy and predictive capability in forecasting POY prices. Based on the model's predictions, we derive the trend of POY prices for 2024, thereby providing a reference for enterprises to formulate optimal material procurement strategies.