Optimization of <scp>CI</scp> engine performance and emission characteristics fueled with cottonseed‐based biodiesel through <scp>ANFIS</scp> and <scp>ANN</scp> approach

Abstract This study reported the application of these soft computing techniques to predict the emissions and performance characteristics of a diesel engine fueled with cottonseed biodiesel under different operating conditions. Fuel injection timing ( o bTDC), fuel injection pressure (bar), biodiesel blend (%), and engine load (%) are the input parameters. The objective of this study was to obtain the finest output for parameters such as brake thermal efficiency (BTE), brake‐specific energy consumption (BSEC), heat release rate (HRR), ignition delay (ID), unburnt hydrocarbons (HC), carbon monoxide (CO), and oxides of nitrogen (NOx). Cottonseed ethyl ester blends of B5, B10, B15, B20, and B25 are employed as fuels. The experiment was conducted using the response surface methodology (RSM) approach. RSM is the ideal combination for improving engine output. Analysis of variance (ANOVA) demonstrated that all of the created models were statistically relevant. Furthermore, three metrics (MSE, RMSE, and R 2 ) are investigated in depth to evaluate the efficacy of soft computing‐based prediction models. In contrast, when it came to forecasting CI engine reactions, both prediction models performed well. Furthermore, it was observed that ANFIS yields more accurate forecast findings than ANN.