| Summary: | 碩士 === 國立成功大學 === 土木工程學系 === 107 === With the rise of the Internet of Things (IoT) technology in recent years, the collection and analysis of a large amount of data have been valued. Thus, the systematic data collection and data analysis have become important issues.
During the building construction process, there are many management forms such as cost, schedule, quality, etc., and most of the data collection relies on the on-site workers to fill out the forms. This kind of operations is labor-intensive, and the amount of data is increasing as the project progresses, so they require an effective data collection system.
This study established an IoT framework for the systematic data collection and thansfer for management information during the construction phase. It developed cloud functions for data networking, computing, and big data analysis. By using the IoT platform (ThingWorx Platform) as the auxiliary software and importing actual construction data, this study described the data transfer connection from the underlying IoT infrastructure to the top-level application layer.
The database of the project management system uses the work package (WP) as the core to summarize and organize the project data, link the database to the IoT platform, create a user interface (UI) that can simulate the sensors, and query the UI for visualizing the project management data for easy browsing.
The quality data of weld assembling are imported into the analysis platform (ThingWorx Analytics). After the defect prediction model was established, the advantages and disadvantages of the seven algorithms compared, the optimal neural network model selected, and the model imported into the ThingWorx Platform, a predicted model UI was established to help with the work arrangement before construction.
|
|---|
