In this study, the effects of fuel blend ratio and engine speed on the performance and emissions of a spark-ignition (SI) engine fueled with gasoline–isopropanol blends were experimentally investigated, modeled, and optimized. Despite the potential benefits of gasoline–alcohol blends for SI engines, many response-surface-based studies adopt simplified surrogate models and fixed second-order formulations, which may not adequately capture coupled and non-linear effects, particularly when experimental data are limited. To address this gap, a data-driven multi-model strategy is adopted to systematically evaluate seven multivariate polynomial regression structures for each response, instead of imposing a single fixed-form model. Model performance is assessed using a hold-out validation scheme, and the model with the highest predictive accuracy is selected for each response as the final predictive model, enabling accurate prediction of torque, fuel consumption (FC), and carbon monoxide (CO), hydrocarbons (HC), and carbon dioxide (CO₂) emissions. As an additional contribution, a multi-objective scalarized function constructed from normalized response values is minimized using a PID-based search algorithm (PSA) to simultaneously maximize torque, power, and brake thermal efficiency (BTE) while minimizing FC, CO, HC, CO₂, and brake-specific fuel consumption (BSFC). The results indicate that increasing the isopropanol ratio has pronounced and non-linear effects on engine behavior: torque and power increase up to an intermediate isopropanol fraction and then decrease at higher ratios, BSFC rises with increasing isopropanol content, and HC, CO, and CO₂ emissions decrease as the isopropanol share increases. The optimization identifies an optimal operating condition at a 50% isopropanol–50% gasoline blend and an engine speed of 2783 rpm. Overall, the study delivers a compact modeling–optimization framework for gasoline–isopropanol operation in SI engines and supports the design of more efficient and environmentally friendly fuel strategies based on alternative fuel usage.
Boğar et al. (Sun,) studied this question.