• Geomechanics and Engineering
• /
• v.20 no.6
• /
• pp.485-495
• /
• 2020

Based on the movement characteristics of overlying strata with gangue backfilling, the compression test of gangue is designed. The deformation characterristics of gangue is obtained based on the different Talbot index. The deformation has a logarithmic growth trend, including sharp deformation stage, linear deformation stage, rheological stage, and the resistance to deformation changes in different stages. The more advantageous Talbot gradation index is obtained to control the surface subsidence. On the basis of similarity simulation test with gangue backfilling, the characteristics of roof failure and the evolution of the supporting force are analyzed. In the early stage of gangue backfilling, beam structure damage directly occurs at the roof, and the layer is separated from the overlying rock. As the working face advances, the crack arch of the basic roof is generated, and the separation layer is closed. Due to the supporting effect of filling gangue, the stress concentration in gangue backfilling stope is relatively mild. Based on the equivalent mining height model of gangue backfilling stope, the relationship between full ratio and mining height is obtained. It is necessary to ensure that the gradation of filling gangue meets the Talbot distribution of n=0.5, and the full ratio meets the protection grade requirements of surface buildings.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.49 no.2
• /
• pp.61-73
• /
• 2021

This study aimed to identify the landscape perception and landscape characteristics of users by utilizing SNS data generated by their experiences. Therefore, how to recognize the main places and scenery appearing on the island, and what are the characteristics of the main scenery were analyzed using online text data and photo data. Text data are text mining and network structural analysis, while photographic data are landscape identification models and color analysis. As a result of the study, First, as a result of frequency analysis of Bakji·Banwol-do topics, we were able to derive keywords for local landscapes such as 'Purple Bridge', 'Doori Village', and location, behavior, and landscape images by analyzing them simultaneously. Second, the network structure analysis showed that the connection between key and undrawn keywords could be more specifically analyzed, indicating that creating landscapes using colors is affecting regional activation. Third, after analyzing the landscape identification model, it was found that artificial elements would be excluded to create preferred landscapes using the main targets of "Purple Bridge" and "Doori Village", and that it would be effective to set a view point of the sea and sky. Fourth, Bakji·Banwol-do were the first islands to be created under the theme of color, and the colors used in artificial facilities were similar to the surrounding environment, and were harmonized with contrasting lighting and saturation values. This study used online data uploaded directly by visitors in the landscape field to identify users' perceptions and objects of the landscape. Furthermore, the use of both text and photographic data to identify landscape recognition and characteristics is significant in that they can specifically identify which landscape and resources they prefer and perceive. In addition, the use of quantitative big data analysis and qualitative landscape identification models in identifying visitors' perceptions of local landscapes will help them understand the landscape more specifically through discussions based on results.