Optimised prediction of tunnel fire heat release rate using the ResNet18_2CLSTM model with bagging for multimodal data
Accurate predictions of HRR will improve preparedness and response strategies, enhance safety, and minimise damage in tunnel fires. In this study, a deep learning prediction model for HRR under multimodal data fusion is proposed. A multimodal dataset is first established to obtain flame images and f...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-11-01
|
| Series: | Case Studies in Thermal Engineering |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2214157X24012991 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846168278710353920 |
|---|---|
| author | Gongyousheng Cui Yuchun Zhang Haowen Tao Shijie Man Haifeng Chen |
| author_facet | Gongyousheng Cui Yuchun Zhang Haowen Tao Shijie Man Haifeng Chen |
| author_sort | Gongyousheng Cui |
| collection | DOAJ |
| description | Accurate predictions of HRR will improve preparedness and response strategies, enhance safety, and minimise damage in tunnel fires. In this study, a deep learning prediction model for HRR under multimodal data fusion is proposed. A multimodal dataset is first established to obtain flame images and flue gas time series data through model-scale tunnel fire experiments. During the model training process, ResNet18 was used to extract features from the flame image, and 2CLSTM was employed to understand the time series of the flame image features and flue gas features to establish the correlation with the HRR. It was evaluated that the error analyses of the measured and predicted values of the validation set yielded R2 greater than 0.85, with errors and standard deviations less than 4 kW. And the model predicted better in the flame growth and decay phases. However, there is some deviation in the predictions near the peak HRR. To address this issue, the Bagging algorithm was introduced to optimise the model. The results show that the ResNet18_2CLSTM model with Bagging reduces the RMSE by 20.47 % and increases the R2 by 4.64 % compared to the original model, and the accuracy is greatly improved. |
| format | Article |
| id | doaj-art-b85177b54608414f959bade15fe039b5 |
| institution | Kabale University |
| issn | 2214-157X |
| language | English |
| publishDate | 2024-11-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Case Studies in Thermal Engineering |
| spelling | doaj-art-b85177b54608414f959bade15fe039b52024-11-14T04:31:48ZengElsevierCase Studies in Thermal Engineering2214-157X2024-11-0163105268Optimised prediction of tunnel fire heat release rate using the ResNet18_2CLSTM model with bagging for multimodal dataGongyousheng Cui0Yuchun Zhang1Haowen Tao2Shijie Man3Haifeng Chen4Department of Fire Protection Engineering, Southwest Jiaotong University, Chengdu, Sichuan, 611756, PR ChinaCorresponding author.; Department of Fire Protection Engineering, Southwest Jiaotong University, Chengdu, Sichuan, 611756, PR ChinaDepartment of Fire Protection Engineering, Southwest Jiaotong University, Chengdu, Sichuan, 611756, PR ChinaDepartment of Fire Protection Engineering, Southwest Jiaotong University, Chengdu, Sichuan, 611756, PR ChinaDepartment of Fire Protection Engineering, Southwest Jiaotong University, Chengdu, Sichuan, 611756, PR ChinaAccurate predictions of HRR will improve preparedness and response strategies, enhance safety, and minimise damage in tunnel fires. In this study, a deep learning prediction model for HRR under multimodal data fusion is proposed. A multimodal dataset is first established to obtain flame images and flue gas time series data through model-scale tunnel fire experiments. During the model training process, ResNet18 was used to extract features from the flame image, and 2CLSTM was employed to understand the time series of the flame image features and flue gas features to establish the correlation with the HRR. It was evaluated that the error analyses of the measured and predicted values of the validation set yielded R2 greater than 0.85, with errors and standard deviations less than 4 kW. And the model predicted better in the flame growth and decay phases. However, there is some deviation in the predictions near the peak HRR. To address this issue, the Bagging algorithm was introduced to optimise the model. The results show that the ResNet18_2CLSTM model with Bagging reduces the RMSE by 20.47 % and increases the R2 by 4.64 % compared to the original model, and the accuracy is greatly improved.http://www.sciencedirect.com/science/article/pii/S2214157X24012991Tunnel fireHeat release rateMultimodal dataDeep learningBagging methods |
| spellingShingle | Gongyousheng Cui Yuchun Zhang Haowen Tao Shijie Man Haifeng Chen Optimised prediction of tunnel fire heat release rate using the ResNet18_2CLSTM model with bagging for multimodal data Case Studies in Thermal Engineering Tunnel fire Heat release rate Multimodal data Deep learning Bagging methods |
| title | Optimised prediction of tunnel fire heat release rate using the ResNet18_2CLSTM model with bagging for multimodal data |
| title_full | Optimised prediction of tunnel fire heat release rate using the ResNet18_2CLSTM model with bagging for multimodal data |
| title_fullStr | Optimised prediction of tunnel fire heat release rate using the ResNet18_2CLSTM model with bagging for multimodal data |
| title_full_unstemmed | Optimised prediction of tunnel fire heat release rate using the ResNet18_2CLSTM model with bagging for multimodal data |
| title_short | Optimised prediction of tunnel fire heat release rate using the ResNet18_2CLSTM model with bagging for multimodal data |
| title_sort | optimised prediction of tunnel fire heat release rate using the resnet18 2clstm model with bagging for multimodal data |
| topic | Tunnel fire Heat release rate Multimodal data Deep learning Bagging methods |
| url | http://www.sciencedirect.com/science/article/pii/S2214157X24012991 |
| work_keys_str_mv | AT gongyoushengcui optimisedpredictionoftunnelfireheatreleaserateusingtheresnet182clstmmodelwithbaggingformultimodaldata AT yuchunzhang optimisedpredictionoftunnelfireheatreleaserateusingtheresnet182clstmmodelwithbaggingformultimodaldata AT haowentao optimisedpredictionoftunnelfireheatreleaserateusingtheresnet182clstmmodelwithbaggingformultimodaldata AT shijieman optimisedpredictionoftunnelfireheatreleaserateusingtheresnet182clstmmodelwithbaggingformultimodaldata AT haifengchen optimisedpredictionoftunnelfireheatreleaserateusingtheresnet182clstmmodelwithbaggingformultimodaldata |