Automatización de la construcción de un ciclo de conducción en base a la recolección de datos de funcionamiento del vehículo en tiempo real

Abstract

The objective of this research was to automate the construction of a driving cycle using different tools and methods to simplify the process, automatically performing calculations and generating a speed-time profile. The data was obtained through OBD-II communication protocols from a data extraction and storage system in the cloud in real-time, developed as a degree project at the Faculty of Informatics and Electronics (FIE). The energy-based micro-trip methodology (EBMT) was employed. It allowed selecting from a set of candidate driving cycles a representative driving cycle with a minimum relative difference concerning the sample. Since four evaluation criteria already established were combined. The monitored data, results, and the driving cycle were displayed and graphed on a web page designed to interact with the user. It was carried out the statistical analysis to verify the effectiveness of the process, comparing the traditional micro-trip methodology with the micro-trip-based method, obtaining favorable results. The automation of the construction process of driving cycles minimized the actions and errors of the researcher. Likewise, the process of generating driving cycles was shorter, allowing more exhaustive analyzes of the characteristics of the driving patterns. It was developed a comparison to corroborate the effectiveness of the methodology, as well as an ANOVA analysis, which yielded satisfactory results for the energy-based micro-trip methodology. Getting a much lower error percentage than that of the traditional methodology. Finally, to obtain a representative driving cycle it is recommended to calculate between 500 and 1000 candidate cycles to minimize the difference with the sample data.