English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Jet fires are relatively small and have the least severe effects among thediverse fire accidents that can occur in industrial plants; however, they areusually involved in a process known as the domino effect, that leads to moresevere events, such as explosions or the initiation of another fire, making theanalysis of such fires an important part of risk analysis. This research workexplores the application of deep learning models in an alternative approachthat uses the semantic segmentation of jet fires flames to extract maingeometrical attributes, relevant for fire risk assessments. A comparison ismade between traditional image processing methods and some state-of-the-artdeep learning models. It is found that the best approach is a deep learningarchitecture known as UNet, along with its two improvements, Attention UNet andUNet++. The models are then used to segment a group of vertical jet flames ofvarying pipe outlet diameters to extract their main geometricalcharacteristics. Attention UNet obtained the best general performance in theapproximation of both height and area of the flames, while also showing astatistically significant difference between it and UNet++. UNet obtained thebest overall performance for the approximation of the lift-off distances;however, there is not enough data to prove a statistically significantdifference between Attention UNet and UNet++. The only instance where UNet++outperformed the other models, was while obtaining the lift-off distances ofthe jet flames with 0.01275 m pipe outlet diameter. In general, the exploredmodels show good agreement between the experimental and predicted values forrelatively large turbulent propane jet flames, released in sonic and subsonicregimes; thus, making these radiation zones segmentation models, a suitableapproach for different jet flame risk management scenarios.

Experimental Large-Scale Jet Flames' Geometrical Features Extraction for Risk Management Using Infrared Images and Deep Learning Segmentation Methods