Non-invasive fuel consumption measurement for internal combustion engines based on Otto cycle

The growing demand for reduced fuel consumption and increased energy efficiency in internal combustion engines underscores the need for continuous advancements in engine technologies. At the same time, the significance of refining driver behavior cannot be ignored, as it stands as a crucial instrument in reducing unnecessary fuel expenses and mitigating escalated levels of pollutant emissions. In this sense, the accurate estimation of fuel consumption is a fundamental prerequisite for the applications of fuel control measures on fuel consumption. Although most modern vehicles offer fuel consumption data assessed by the electronic central unit, this information is primarily designed to provide users with an estimate of their momentary average consumption, without any assurance from manufacturers regarding the reliability of this data. Most of the initiatives aimed at measuring fuel consumption require invasive approaches, and there is still a lack of studies addressing non-invasive methods to evaluate the total fuel consumption, especially for Otto cycle engines operating with pure ethanol and a mixture of gasoline and ethanol. In this regard, the objective of this work is to develop a non-invasive method for the real-time fuel measurement of internal combustion engines based on the Otto cycle with a common rail fuel injection system. The proposed technique requires only the measurement of the electrical signal pulses sent from the engine control unit to the fuel injector and does not affect the performance or operation of the engine in any way. The measurement system was built using low-cost electronic components, and its accuracy was evaluated using a single-cylinder research engine (SCRE). A series of 32 tests were performed, considering four different engine loads, four different speeds, and two different fuels, ethanol (E100) and gasoline and ethanol blend (E27). The results achieved were superior to those obtained with electromechanical sensors. The results obtained by measuring the fuel consumption with the proposed methodology showed a maximum percentage error of ± 2.85% for ethanol (E100) and ± 3.30% for a blend of gasoline with 27% ethanol (E27).