Experimental and predictive analysis of performance, emission, and combustion of a heavy-duty HCNG fueled spark-ignition engine by optimized support vector machine

Hydrogen, as a secondary fuel, offers significant potential for improving the performance and emissions of spark-ignition (SI) engines. This study aims to evaluate and optimize the performance, emissions, and combustion parameters of a hydrogen-enriched compressed natural gas (HCNG) fueled engine under stoichiometric conditions. Experiments were conducted by varying exhaust gas recirculation (EGR) rates (3 %–29 %), spark timing (11°–49° CA bTDC), engine speed (1200 rpm and 1700 rpm), and load (low to high) for HCNG20 % fuel. The results demonstrated a 37.12 % increase in brake thermal efficiency (BTE) at 27° CA bTDC and a peak torque of 1519.346 Nm at 29° CA bTDC under high load at 1200 rpm, achieved by raising the EGR rate from 24 % to 29 %. A reduction of 73.15 % in NOx emissions and 46.69 % in CO emissions was observed at 47° CA bTDC at 1200 rpm by varying the EGR rate from 20 % to 25 %. Total hydrocarbon (THC) emissions increased by 34.35 % under the aforementioned operating conditions. Peak in-cylinder pressure, temperature, and heat release rate (HRR) decreased by 21.66 %, 8.95 %, and 22.01 %, respectively, and the crank angle shifted away from top dead center (TDC) from 368° to 392° CA bTDC by increasing the EGR rate from 3 % to 12 % at medium load. Furthermore, performance and emissions were predicted using support vector machine (SVM). The Medium Gaussian SVM (MGSVM) achieved significant reductions in root mean square error (RMSE) up to 53.98 % for BTE, 23.13 % for NOx emissions, and 26.58 % for CO emissions under medium load conditions. These findings provide valuable insights for optimizing engine performance and emissions in heavy-duty HCNG-fueled engines, contributing to cleaner and more efficient transportation solutions.