Diagnóstico de Emisiones Vehiculares Aplicando Lógica Difusa

In the operation of a spark ignition engine, the variables associated with the combustion process present a certain degree of uncertainty that, in one way or another, influence maintenance actions. Among the main causes of excessive pollutant emissions are incorrect air-fuel ratio and poor ignition timing, which in turn alter the composition of the exhaust gases. This leads to errors in the estimation of the lambda factor by gas analysis equipment. To avoid this problem, a diagnostic system based on fuzzy logic was developed that estimates the values of the lambda factor and ignition timing. The system takes as inputs the concentrations of carbon monoxide (CO), unburned hydrocarbons (HC), carbon dioxide (CO2), and oxygen (O2). The rules were developed from the results of an experimental design from which the relationships between the input and output variables are derived. To validate the resulting system, the results of measurements are taken with the engine under various operating conditions, altering its timing.