• Steel and Composite Structures
• /
• v.46 no.1
• /
• pp.53-73
• /
• 2023

Man-made structure materials like concrete usually contain inclusions. These inclusions affect the mechanical properties of concrete. In this investigation, the influence of inclusion length and inclination angle on three-dimensional failure mechanism of concrete under uniaxial compression were performed using experimental test and numerical simulation. Approach of acoustic emission were jointly used to analyze the damage and fracture process. Besides, by combining the stress-strain behavior, quantitative determination of the thresholds of crack stress were done. concrete specimens with dimensions of 120 mm × 150 mm × 100 mm were provided. One and two holes filled by gypsum are incorporated in concrete samples. To build the inclusion, firstly cylinder steel tube was pre-inserting into the concrete and removing them after the initial hardening of the specimen. Secondly, the gypsum was poured into the holes. Tensile strengths of concrete and gypsum were 2.45 MPa and 1.5 MPa, respectively. The angle bertween inclusions and axial loadind ary from 0 to 90 with increases of 30. The length of inclusion vary from 25 mm to 100 mm with increases of 25 mm. Diameter of the hole was 20 mm. Entirely 20 various models were examined under uniaxial test. Simultaneous with experimental tests, numerical simulation (Particle flow code in two dimension) were carried out on the numerical models containing the inclusions. The numerical model were calibrated firstly by experimental outputs and then failure behavior of models containing inclusions have been investigated. The angle bertween inclusions and axial loadind vary from 0 to 90 with increases of 15. The length of inclusion vary from 25 mm to 100 mm with increases of 25 mm. Entirely 32 various models were examined under uniaxial test. Loading rate was 0.05 mm/sec. The results indicated that when inclusion has occupied 100% of sample thickness, two tensile cracks originated from boundaries of sample and spread parallel to the loading direction until being integrated together. When inclusion has occupied 75% of sample thickness, four tensile cracks originated from boundaries of sample and spread parallel to the loading direction until being integrated together. When inclusions have occupied 50% and 25% of sample thickness, four tensile cracks originated from boundaries of sample and spread parallel to the loading direction until being integrated together. Also the inclusion was failed by one tensile crack. The compressive strength of samples decease with the decreases of the inclusions length, and inclusion angle had some effects on that. Failure of concrete is mostly due to the tensile crack. The behavior of crack, was affected by the inclusion length and inclusion number.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_3
• /
• pp.979-995
• /
• 2023

As wildfires are difficult to predict, real-time monitoring is crucial for a timely response. Geostationary satellite images are very useful for active fire detection because they can monitor a vast area with high temporal resolution (e.g., 2 min). Existing satellite-based active fire detection algorithms detect thermal outliers using threshold values based on the statistical analysis of brightness temperature. However, the difficulty in establishing suitable thresholds for such threshold-based methods hinders their ability to detect fires with low intensity and achieve generalized performance. In light of these challenges, machine learning has emerged as a potential-solution. Until now, relatively simple techniques such as random forest, Vanilla convolutional neural network (CNN), and U-net have been applied for active fire detection. Therefore, this study proposed an active fire detection algorithm using state-of-the-art (SOTA) deep learning techniques using data from the Advanced Himawari Imager and evaluated it over East Asia and Australia. The SOTA model was developed by applying EfficientNet and lion optimizer, and the results were compared with the model using the Vanilla CNN structure. EfficientNet outperformed CNN with F1-scores of 0.88 and 0.83 in East Asia and Australia, respectively. The performance was better after using weighted loss, equal sampling, and image augmentation techniques to fix data imbalance issues compared to before the techniques were used, resulting in F1-scores of 0.92 in East Asia and 0.84 in Australia. It is anticipated that timely responses facilitated by the SOTA deep learning-based approach for active fire detection will effectively mitigate the damage caused by wildfires.