Optimization of performance and emission parameters of direct injection diesel engine fuelled with microalgae Spirulina (L.) - Response surface methodology and full factorial method approach

The closeness of the characteristics of biodiesel with that of diesel during combustion can be achieved through variations in compression ratios of the engine coupled with different biodiesel blends. The simplest approach towards attaining an optimal combination can be achieved through design of experiments (DOE). The present study has been greatly focussed on the usage of blends of microalgae Spirulina (L.) with conventional diesel in a variable compression ratio engine at varying loads, compression ratios (CR) and concentration of blends. L64 orthogonal array using full factorial method was used in designing the experiments and the corresponding multiple responses were studied using response surface methodology (RSM). The effects of load, CR and blends along with the correlation between the performance parameters (brake specific fuel consumption and brake thermal efficiency) and emission parameters (carbon dioxide, particulate matter and oxides of nitrogen) were studied. The model validation has been performed using the adjusted value of R-2, which resulted in the decrease of unnecessary terms in the designed mode. The regression model explained the mechanism by delivering the response variable of R-2 by more than 80%. The model predicted that the load of 64.634, CR of 16.50 and 20% blend would provide BTE of 31.357% and BSFC of 274.97 g/kWh with lower values of CO2 (869.075 g/kWh), PM (0.2807 g/kWh) and NOx (1804.97 ppm), respectively. The confirmation test showed the closeness of the predicted results with those of experimental values.

Automated summary

This study focused on investigating the combustion characteristics of microalgae Spirulina blends with conventional diesel in a variable compression ratio engine. The effects of load, compression ratio, and blends on performance parameters and emission parameters were studied. The model validation and confirmation tests showed that the predicted results were close to the experimental values.