Uma aplicação para carga de dados de monitoramento da geometria de linhas férreas

Abstract

Along the advancement of technology, new ways of monitoring sensor data are being proposed. In this work, we intend to explore a solution to load, store, retrieve and delete data that reflects the state of railroad tracks, which are collected from sensors installed in a control car. Initially, the nature of the data is analyzed and, considering that it is a series of temporal data, we set out to search for and test an alternative database management system. Thus, the tool "TimescaleDB" was found, which was promising initially, so to increase the performance of the operations of registration, removal and recovery of large volumes of temporal data. In addition, traditional tools for storing data, such as PostgreSQL, in versions 9.6 and 10 were explored. From its tenth version, an instrument to deal with this specific data type was made available, and it was also considered. After this, tests were conducted from the proposed solution, but did not show satisfactory results regarding the use of TimescaleDB, which led to the non-use of the tool.