• Geophysics and Geophysical Exploration
• /
• v.27 no.3
• /
• pp.171-180
• /
• 2024

The processing of seismic data involves analyzing earthquake wave data to understand the internal structure and characteristics of the Earth, which requires high computational power. Recently, machine learning (ML) techniques have been introduced to address these challenges and have been utilized in various tasks such as noise reduction and velocity model construction. However, most studies have focused on specific seismic data processing tasks, limiting the full utilization of similar features and structures inherent in the datasets. In this study, we compared the efficacy of using receiver-wise time-series data ("receiver array") and synchronized receiver signals ("time array") from shotgathers for pretraining a Bidirectional Encoder Representations from Transformers (BERT) model. To this end, shotgather data generated from a synthetic model containing faults was used to perform noise reduction, velocity prediction, and fault detection tasks. In the task of random noise reduction, both the receiver and time arrays showed good performance. However, for tasks requiring the identification of spatial distributions, such as velocity estimation and fault detection, the results from the time array were superior.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.11
• /
• pp.110-117
• /
• 2017

In the construction of high-rise residential buildings in which the finishing works are done repetitively in each housing unit, the construction period and cost can be reduced, while maintaining the quality, if the work process is properly managed in order for the resources to be input continuously. Time management methods such as the Bar Chart, CPM and LOB are not appropriate, due to the difficulties involved in the diagraming of repetitive works and applying them to the project. The Tact method also has difficulties in maintaining a constant work flow and needs significant effort and cooperation from the subcontractors to allocate the resources consistently. Partitioning, base works for finishing, floor mortar plastering and the final finishing work are done sequentially in residential buildings projects, and there are many repetitive activities which differ in terms of the work method, work area and productivity. If these repetitive activities are synchronized or converged toward the last work area, the goal of process management can be achieved effectively. Therefore, a process planning method for the finishing works of residential building projects is proposed, which takes into consideration elements, such as the sequential relation between the activities in each housing unit and classification of repetitive works in terms of their management method, work area and production rate, for the continuous input of resources into the housing units.