Polynomial regression model for the prediction of thermal efficiency in spark ignition engines operating with ethanol-water mixtures

The present work aims to propose a polynomial model 𝑦= 𝑃(𝑉1,𝑉2,…,𝑉𝑛)capable of predicting the thermal efficiency [𝑦] of a single cylinder spark ignition reciprocating engine that uses ethanol and water mixed as fuel at various proportions and speeds. The independent variables used in the development of the polynomial model are: 𝑉1= water fraction, 𝑉2= rotation speed [rpm], 𝑉3=compression ratio, � �4=ignition advance angle [°BTDC], 𝑉5 =fuel temperature [°C], � �6=exhaust temperature [°C], 𝑉7=oil temperature [°C], 𝑉8=engine torque [Nm], 𝑉9=fuel consumption [g/s] and 𝑉10=power [kW] and 𝑉11=specific fuel consumption [g/kWh]. A determination coefficient 𝑅2= 0.9785 was obtained, which indicates that the polynomial model obtained can predict the behavior of thermal efficiency with an accuracy of 97.85%. The database used in this work is from the authors Torres, Mendes and Alburquerque [2024].