Coupling of green building construction based on particle Swarm optimizing neural network algorithm
In the continuous development of the green building industry, construction safety management faces increasing challenges, particularly in safety and environmental protection, which requires precise evaluation and control. Therefore, this study proposes a coupling analysis method for green building c...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
EDP Sciences
2025-01-01
|
| Series: | Sustainable Buildings |
| Subjects: | |
| Online Access: | https://www.sustainable-buildings-journal.org/articles/sbuild/full_html/2025/01/sbuild20240019/sbuild20240019.html |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In the continuous development of the green building industry, construction safety management faces increasing challenges, particularly in safety and environmental protection, which requires precise evaluation and control. Therefore, this study proposes a coupling analysis method for green building construction based on particle swarm optimisation neural network. The purpose is to strengthen safety risk management in green building construction by combining particle swarm optimisation with neural network algorithms. A risk coupling performance comparison was conducted between traditional and research algorithms. In the results, when using a back propagation neural network for prediction, the actual construction risk rate increased from 0.235 to 0.431. the optimised algorithm showed an increase from 0.168 to 0.453, and the prediction error improving from −0.352 to 0.014, demonstrating a high degree of adaptability and accuracy to actual changes. Compared with traditional methods, the prediction error of this algorithm is significantly reduced, and the data fitting accuracy is improved to 0.99809, indicate its effectiveness in predicting construction safety risks. The research results not only contribute to improving the efficiency of safety management during the construction process, but also provide technical support for risk prediction models in the future green building field. |
|---|---|
| ISSN: | 2492-6035 |