Tesis
Elaboración de un algoritmo para la construcción de ciclos de conducción
Fecha
2021-11-25Registro en:
Plaza Guerrero, José Manuel. (2021). Elaboración de un algoritmo para la construcción de ciclos de conducción. Escuela Superior Politécnica de Chimborazo. Riobamba.
Autor
Plaza Guerrero, José Manuel
Resumen
In the present investigation it is proposed to obtain driving cycles through the use of an algorithm written in Python developed in the IDE spyder; establishing a new methodology VSPBMT (Specific Vehicular Speed Based on Micro Trips) to obtain driving cycles, which consists of dividing a database (the values established every second during trips are recorded) in micro trips, it was used four parameters as cycle evaluation criteria: average speed, positive average acceleration, idle time percentage and vehicle specific power. This last parameter was calculated by adding the forces that oppose the movement of the vehicle, multiplied by the speed and divided by the weight of the vehicle; this parameter is calculated for each second of the entire trip. This new VSPBMT methodology will be compared with the EBMT methodology (Energy Based on Micro Trips), which has been developed in a previous research carried out at the Tecnológico de Monterrey. After correctly filtering the data obtained from the previous investigation, the analysis of these data was carried out using the two methodologies (algorithm); for which it is recommended to carry out a thousand iterations in each sample to obtain more exact results. Ten samples (cycles and characteristic parameters) were taken with each methodology and a statistical analysis (Anova, Variance, Fisher) was performed comparing the means of the ten cycles of each method and obtaining results that demonstrate the similarity of both methods; thus ensuring the effectiveness of the VSPBMT methodology for obtaining representative driving cycles for a given city or region. However, it was shown that with the EBMT methodology, more accurate driving cycles are obtained, due to the fact that a relatively smaller error is generated, unlike the VSPBMT methodology.